OpenSeesPy Documentation

Release 3.5.1.3

Minjie Zhu

Jun 08, 2023

 Contents

1 Developer 3

Bibliography 439

Index 441

i
ii
 OpenSeesPy Documentation, Release 3.5.1.3

Important: Version 3.5.1.3 is released!

Python 3.10 is required for Windows.
OpenSeesPy is on PyPi (Windows, Linux, Mac).
OpenSeesPy is free for research, education, and internal use. Commercial redistribution of OpenSeesPy, such as,
but not limited to, an application or cloud-based service that uses import openseespy, requires a license similar to
that required for commercial redistribution of OpenSees.exe. Contact Dr. Minjie Zhu ([email protected]) for
commercial licensing details.

Note: OpenSees Cloud is a SaaS implementation of OpenSees using cloud computing. Analyses run in the cloud on
AWS while the front end integrates input, output, documentation, and a model viewer.
OpenSees Amazon Machine Image is a virtual machine that runs OpenSeesPy and OpenSeesMP with no additional
compilation or installation required at low cost.

Note: Questions including modeling issues and the use of OpenSeesPy, please post on OpenSeesPy Forum.
You are very welcome to contribute to OpenSeesPy with new command documents and examples by sending pull
requests through github pulls.
For errors in this document, submit on github issues.

OpenSeesPy is a Python 3 interpreter of OpenSees. A minimum script is shown below:

# import OpenSeesPy
import openseespy.opensees as ops

# wipe model
ops.wipe()

# create model
ops.model('basic', '-ndm', 2, '-ndf', 3)

# print model
ops.printModel()

Contents 1
OpenSeesPy Documentation, Release 3.5.1.3

2 Contents
 CHAPTER 1

Developer

Minjie Zhu

Research Associate
Civil and Construction Engineering
Oregon State University

1.1 Installation

1. PyPi (Windows, Linux, Mac)

2. Other Installation

1.1.1 PyPi (Windows, Linux, Mac)

1. PyPi (Windows)
2. PyPi (Linux)
3. PyPi (Mac)

PyPi (Windows)

Install Anaconda

• Install Anaconda

3
OpenSeesPy Documentation, Release 3.5.1.3

Install In Jupyter Notebook

• Start Jupyter Notebook

• run install command in the notebook

Install In command line

• Start Anaconda Prompt

4 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

• To install

python -m pip install openseespy

python -m pip install --user openseespy

• To upgrade

python -m pip install --upgrade openseespy

python -m pip install --user --upgrade openseespy

• To import

import openseespy.opensees as ops

PyPi (Linux)

Install Python

• Install Python 3 using your Linux’s package manager

– Ubuntu/Debian: sudo apt install python3 python3-pip
– Centos/Redhat: sudo yum install python3 python3-pip

Install in terminal

• To install

python3 -m pip install openseespy

python3 -m pip install --user openseespy

• To upgrade

1.1. Installation 5
OpenSeesPy Documentation, Release 3.5.1.3

python3 -m pip install --upgrade openseespy

python3 -m pip install --user --upgrade openseespy

• To import
import openseespy.opensees as ops

PyPi (Mac)

Install Python

• Install Python 3.8

Install in terminal

• To install
python3.8 -m pip install openseespy

python3.8 -m pip install --user openseespy

• To upgrade
python3.8 -m pip install --upgrade openseespy

python3.8 -m pip install --user --upgrade openseespy

• To import
import openseespy.opensees as ops

1.1.2 Other Installation

1. DesignSafe (Web-based)
2. Windows Subsystem for Linux (Windows)

DesignSafe (Web-based)

OpenSeesPy has been offical in DesignSafe.

1. OpenSeesPy in DesignSafe
2. Paraview in DesignSafe

OpenSeesPy in DesignSafe

Follow steps below to run OpenSeesPy in DesignSafe.

Tip:

6 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

• Go To DesignSafe website and click register

Tip:
• Register an account for free and log in

1.1. Installation 7
OpenSeesPy Documentation, Release 3.5.1.3

Tip:
• Land on your own portal and go to workspace

8 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

Tip:
• In your workspace, select Jupyter, launch it, and start my server

1.1. Installation 9
OpenSeesPy Documentation, Release 3.5.1.3

Tip:
• Now you should be in the Jupyter Notebook
• Go to mydata and alwasy save your data under this folder

10 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

Tip:
• In the mydata folder, select New and then Python 3

Tip:
• The OpenSeesPy version on DesignSafe is not the latest.
• To update to the latest and import:

1.1. Installation 11
OpenSeesPy Documentation, Release 3.5.1.3

Tip:
• Now you can write OpenSeesPy script in Jupyter Notebook, run the script, and show results
• To show figure in the Notebook, you should include %matplotlib inline at the beginning

12 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

Paraview in DesignSafe

DesignSafe provides paraview for viewing OpenSeesPy results:

Tip:
• Make sure the steps in OpenSeesPy in DesignSafe are completed.

Tip:
• Go to Workspace and select Visualizaion and Paraview.

Tip:
• In the Job Submission windows, you should select Working Directory, Maximum job runtime, Job
name, and Node Count, and click Run.

1.1. Installation 13
OpenSeesPy Documentation, Release 3.5.1.3

Tip:
• You can see the job status on the right

14 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

Tip:
• Wait until see this windows and connect to Paraview

1.1. Installation 15
OpenSeesPy Documentation, Release 3.5.1.3

Tip:
• Now open the pvd file in Paraview

16 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

Tip:
• Initially, you see nothing
• Check the Pipeline Browser
• Click on the eye left to the file in Pipeline Browser

Tip:
• Change the Solid Color to other variables
• Change Surface to Surface With Edges

1.1. Installation 17
OpenSeesPy Documentation, Release 3.5.1.3

Windows Subsystem for Linux (Windows)

This is a real Linux subsystem for you to run OpenSeesPy Linux version on Windows.

Install the Windows Subsystem for Linux

Follow the instruction on here to install Windows Subsystem for Linux on Windows 10. There are a couple of Linux
distributions available and Ubuntu is recommended. Once the Linux is installed, it will show as an application in the
start menu.

18 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

Install Anaconda and start Jupyter Notebook

• Run the subsystem from start menu and a terminal window will show.

• Download Anaconda Linux version with command

~$ wget https://repo.anaconda.com/archive/Anaconda3-2021.05-Linux-x86_64.sh

• Install Anconda Linux version with commands

~$ bash Anaconda3-2021.05-Linux-x86_64.sh

>>> Please answer 'yes' or 'no':

>>> yes

>>> Anaconda3 will not be installed into this location:

[/home/username/anaconda3] >>> (enter)

• Start Jupyter Notebook

~$ /home/username/anaconda3/bin/jupyter-notebook

• Copy the address in red box to a web browser

1.1. Installation 19
OpenSeesPy Documentation, Release 3.5.1.3

In Jupyter Notebook

Start a new notebook and then

20 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

In the command line (optional)

• Run Anaconda with following command, where username is your username of your computer. Please use the
username shown in last step

~$ /home/username/anaconda3/bin/python3.8

• Install or Upgrade OpenSeesPy with commands

~$ /home/username/anaconda3/bin/python3.8 -m pip install openseespy

~$ /home/username/anaconda3/bin/python3.8 -m pip install --upgrade openseespy

1.1. Installation 21
OpenSeesPy Documentation, Release 3.5.1.3

• Run OpenSeesPy
First run Anaconda with

~$ /home/username/anaconda3/bin/python3.8

Then import OpenSeesPy with

import openseespy.opensees as ops

ops.printModel()

• run OpenSeesPy scripts

/home/username/anaconda3/bin/python3.8 script.py

1.2 Compilation

1. Compilation for Windows

2. Compilation for MacOS
3. Compilation using QT

22 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

1.2.1 Compilation for Windows

1.2.2 Compilation for MacOS

This is a suggestive guide for MacOS.

• Download OpenSees Source code

git clone https://github.com/OpenSees/OpenSees.git

• Install MacPorts for your macOs.

• Install gcc8

sudo port install gcc8

• Install Python 3.8 using MacPorts

sudo port install python38 python38-devel

• Install boost using Macports

sudo port install boost

• Create Makefile.def
– Copy one of MAKES/Makefile.def.MacOS10.x to the root and rename to Makefile.def
– Change the path and variables for your system
• Compile
– Run make python -j from root
– The library file will be at SRC/interpreter/opensees.so

1.2.3 Compilation using QT

Following is the OpenSees Compilation using QT by

Stevan Gavrilovic github

PhD Candidate
University of British Columbia

Original instructions can be found at here.

A Qt build environment for OpenSees

OpenSeesQt
A Qt build environment for OpenSees – Open System For Earthquake Engineering Simulation Pacific Earthquake
Engineering Research Center(http://opensees.berkeley.edu).
Qt is an open source, cross-platform development environment that is free for many uses. Please see the license.

1.2. Compilation 23
OpenSeesPy Documentation, Release 3.5.1.3

Dependency

The purpose of this project is to create a package that will allow aspiring developers to get started on writing code
without having to worry about the compilation environment. A program as large as OpenSees relies on many third-
party libraries, often referred to as dependencies. It can be a daunting task assembling, compiling, and linking these
libraries. Many times, these libraries depend on other libraries, and so on. The current list of dependencies includes:
• MUMPS (5.1.2)
• Scalapack (2.0.2.13)
• UMFPACK (5.7.7) contained in Suite-Sparse (5.3.0)
• SUPERLU (5.2.1)
• SUPERLUMT (3.0)
• SUPERLUDIST (5.1.0)
• Openblas (0.3.5)
• Parmetis (4.0.3)
• ARPACK (3.6.3)
• Libevent (2.1.8)
• GCC(8.2.0)
• TCL (8.6.9)** Tcl only
• Python(3.7.2)** OpenSeesPy only
Please ensure that your project complies with each library’s licensing requirements.
Another feature of this build platform is modularity. Developers can select from a list of build options that can be
turned on and off as required. The basic configuration builds the structural analysis core. Other build options include
parallel processing, reliability, and the particle finite element method (PFEM) modules. Python and Tcl interpreters
are additional build options. These options are located in single configuration file called qmake.conf. By default, the
environment is configured to build the core along with the parallel processing module. Other options can be turned on
by deleting the ‘#’symbol that precedes the option; this includes the option into the build environment.

Note: Note: This build environment comes with pre-compiled dependencies. Although this makes getting started
easier, the caveat is that your build environment (compiler version) must match that of the environment used to compile
the dependencies. The supported environments are listed below. There are no guarantees that it will work with other
build environments. In other words, make sure your compiler type and version (i.e., clang-1000.11.45.5) matches the
version below listed under the heading ‘Supported Build Environments’. Otherwise, bad things might happen. Also,
this project is still a work in progress. Currently, only building in OS X is supported. Windows support will be added
shortly. Moreover, not all build options are supported. For example, compiling with fortran is not supported.

Note: This project uses qmake and Qt Creator. qmake is a build tool for compiling and linking applications. Qt
Creator is a free IDE (interactive development environment) that bundles code writing/editing and application building
within one program. Qt Creator uses project (.pro) files. The project files contain all information required by qmake
to build an application.

24 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

Getting started:

1. Download and install Qt open source from https://www.qt.io/download The version of the Qt library is not
important in this case since the library is not used in the OpenSees project (although I use Qt in other projects
and recommend it)
2. Download the OpenSeesQt source code into a folder of your choice (https://github.com/steva44/OpenSees/
archive/master.zip)
3. In the directory containing the code, double click on the ‘OpenSees.pro’ file. If compiling OpenSeesPy, open
the ‘OpenSeesPy.pro’ file. The .pro files are project files that will automatically open the project in Qt Creator.
4. Select a build configuration, for example ‘Desktop Qt 5.12.1 clang 64bit’. The project will automatically con-
figure itself. You only have to do this once.
5. The left-hand pane should now display the project directory structure. In the left-hand pane, under the heading
qmake, open the ‘qmake.conf’ file. Review and select additional build options, if any. Note that this is still a
work in progress and not all build options are supported.
6. Click on the ‘start’ button in the bottom lefthand corner of Qt Creator to compile. Clicking on the small computer
symbol above the start button allows for switching between the debug and release deploy configurations. The
release deployment results in faster program execution but it does not allow for debugging or stepping through
the code. The start button with the bug symbol opens the debugger.
7. Go and have a coffee, it will take a few minutes to finish compiling!

Building OpenSeesPy:

OpenSeesPy builds OpenSees as a library object that can be used within Python.
Steps: Follow steps 1-4 under the heading getting started above.
1. The left-hand pane should now display the project directory structure. In the left-hand pane, under the heading
qmake, open the ‘qmake.conf’ file. Under the heading #INTERPRETERS, uncomment the _PYTHON op-
tion by removing the ‘#’ symbol. Everything else should be configured automatically going forward. Python
automatically compiles with the reliability, parallel, and PFEM modules.
2. The last few lines at the end of the ‘OpenSeesPy.pro’ file contain the location of the Python framework. Update
this so that it matches the location of Python on your build system.
3. Click on the ‘start’ button in the bottom lefthand corner of Qt Creator to start compiling. Clicking on the
small computer symbol allows for switching between the debug and release deploy configurations. The release
deployment results in faster program execution but it does not allow for debugging or stepping through the code.
Build in release mode if using OpenSees as a library in a Python project.
4. Go and have a coffee, it will take a few minutes to finish compiling!
5. After successful compilation, the library will be in the ‘bin’ folder. The bin folder is located in the ‘build’ folder
which is created, by default, one directory higher than the OpenSeesQt source code. The name of the build
folder should look something like this: build-OpenSeesPy-Desktop_Qt_5_12_1_clang_64bit-Debug
6. OS X only
OS X automatically prepends a ‘lib’ to the library file. Remove this ‘lib’ and rename the file to be
‘opensees.dylib’ Next, a symbolic link is required for a Python project to import the library. To create a symbolic
link, cd the directory containing the OpenSees library in terminal and run the following command to create a
symbolic link:

ln -s opensees.dylib opensees.so

1.2. Compilation 25
OpenSeesPy Documentation, Release 3.5.1.3

There should now be a .so (shared object) file in addition to the .dylib file. Finally, copy both the .dylib and the .so
‘link’ into your python environment folder to import it into your project. Directions for using OpenSeesPy can be
found at the project website: https://openseespydoc.readthedocs.io/en/latest/index.html

Supported Build Environments:

OSX
Build Environment:
• OSX 10.14.3 (Mojave)
• Qt 5.12.1
• Qt Creator 4.8.1
Compiler:
• Apple LLVM version 10.0.0 (clang-1000.11.45.5)
• Target: x86_64-apple-darwin18.2.0
• Thread model: posix 64-BIT architecture
To find the version of clang on your computer, type the following in terminal:

clang --version

Note: This project comes with pre-built libraries for everything except Python. Therefore, you do not have to go
through the trouble of building any libraries unless you are using a special build system or you want to experiment.
The precompiled library files are located in the ‘OpenSeesLibs’ folder. In the event that you are feeling adventurous
and you want to compile the libraries on your own, instructions are given below for each library, for each operating
system. After successful compilation, note the installation directory. This directory contains the locations of the
‘include’ and ‘lib’ folders for that library. If replacing or adding new libraries, the file paths should be updated in
the ‘OpenSeesLibs.pri’ file. This is required so that the compiler knows where to find the header files and to link the
libraries to your project.

OSX
On OSX, the dependencies are built/installed with Homebrew. Homebrew is a free and open-source software pack-
age management system that simplifies the installation of software on Apple’s macOS operating system and Linux.
Homebrew maintains its own folder within /usr/local/ directory aptly named the 'Cellar':

/usr/local/Cellar/

Each dependency installed through Homebrew will have its own subfolder within the Cellar directory. Each subfolder
contains that dependencies 'include' and 'lib' folders.

MUMPS

MUltifrontal Massively Parallel sparse direct Solver, or MUMPS, is a sparse direct solver used for parallel solving of
a system of equations
Installing MUMPS via brew: Dominique Orban has written a Homebrew formula (http://brew.sh) for Mac OSX users.
Homebrew MUMPS is now available via the OpenBLAS tap. Build instructions are as follows:
In terminal, copy and paste each command individually and execute:

26 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

brew tap dpo/openblas

brew tap-pin dpo/openblas
brew options mumps # to discover build options
brew install mumps [options...]

The options can be left blank, i.e., with default options so the last line will look like:

brew install mumps

Mumps requires the following dependencies that will automatically be installed:

-Scalapack

OpenMPI

OpenMPI is a high performance message passing library (https://www.open-mpi.org/)

Installing OpenMpi via brew: In terminal, copy and paste the following command and execute:

brew install open-mpi

OpenMPI requires the following dependencies that will automatically be installed:

• GCC (GNU compiler collection)
• libevent (Asynchronous event library: https://libevent.org/)

UMFPACK

UMFPACK is a set of routines for solving unsymmetric sparse linear systems of the form Ax=b, using the Unsym-
metric MultiFrontal method (Matrix A is not required to be symmetric). UMFPACK is part of suite-sparse library in
homebrew/science
In terminal, copy and paste each command individually and execute:

brew tap homebrew/science

brew install suite-sparse

UMFPACK requires the following dependencies that will automatically be installed:

• Metis (‘METIS’ is a type of GraphPartitioner and numberer - An Unstructured Graph Partitioning And Sparse
Matrix Ordering System’, developed by G. Karypis and V. Kumar at the University of Minnesota.

SUPERLU

SUPERLU is a general purpose library for the direct solution of large, sparse, nonsymmetric systems of linear equa-
tions. The library is written in C and is callable from either C or Fortran program. It uses MPI, OpenMP and CUDA
to support various forms of parallelism.
Installing SUPERLU via brew In terminal, copy and paste the following command and execute:

brew install superlu

Should install by default with option --with-openmp enabled. Open MP is needed for parallel analysis.
SUPERLU requires the following dependencies that will automatically be installed:

1.2. Compilation 27
OpenSeesPy Documentation, Release 3.5.1.3

• GCC (GNU compiler collection)

• openblas (In scientific computing, OpenBLAS is an open source implementation of the BLAS API with many
hand-crafted optimizations for specific processor types)

SUPERLUMT

SUPERLU but for for shared memory parallel machines. Provides Pthreads and OpenMP interfaces.
Installing SUPERLUMT via brew: In terminal, copy and paste the following command and execute:

brew install superlu_mt

SUPERLUMT requires the following dependencies that will automatically be installed:

• openblas

SUPERLUDIST

SUPERLU but for for for distributed memory parallel machines. Supports manycore heterogeous node architecture:
MPI is used for interprocess communication, OpenMP is used for on-node threading, CUDA is used for computing on
GPUs.
Installing SUPERLUDIST via brew: In terminal, copy and paste the following command and execute:

brew install superlu_dist

SUPERLUDIST requires the following dependencies that will automatically be installed:

• GCC (GNU compiler collection)
• openblas (In scientific computing, OpenBLAS is an open source implementation of the BLAS API with many
hand-crafted optimizations for specific processor types)
• OpenMPI (a high performance message passing library (https://www.open-mpi.org/))
• Parmetis (MPI library for graph/mesh partitioning and fill-reducing orderings)

LAPACK (SCALAPACK)

The Linear Algebra PACKage, or LAPACK, is written in Fortran 90 and provides routines for solving systems of
simultaneous linear equations, least-squares solutions of linear systems of equations, eigenvalue problems, and singu-
lar value problems.The associated matrix factorizations (LU, Cholesky, QR, SVD, Schur, generalized Schur) are also
provided, as are related computations such as reordering of the Schur factorizations and estimating condition numbers.
Dense and banded matrices are handled, but not general sparse matrices. In all areas, similar functionality is provided
for real and complex matrices, in both single and double precision.
LAPACK is given as a system library in OSX, you may have to update the locations of your system library in
‘OpenSeesLibs.pri’

BLAS

The BLAS (Basic Linear Algebra Subprograms) are routines that provide standard building blocks for performing
basic vector and matrix operations.
BLAS is given as a system library in OSX, you may have to update the locations of your system library in
‘OpenSeesLibs.pri’

28 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

ARPACK

ARPACK contains routines to solve large scale eigenvalue problems

Installing ARPACK via brew: In terminal, copy and paste the following command and execute:

brew install arpack

ARPACK requires the following dependencies that will automatically be installed:

• GCC (GNU compiler collection)
• openblas (In scientific computing, OpenBLAS is an open source implementation of the BLAS API with many
hand-crafted optimizations for specific processor types)

GCC

Many of the dependencies require fortran (there is still a lot of legacy fortran code floating around in the engineering
world). On OSX, I found the best solution is to use the pre-bundled fortran capabilities in the GNU compiler collection
or GCC. In addition to its fortran capabilities, GCC is a dependency for many other libraries.
Installing GCC via brew: In terminal, copy and paste the following command and execute:

brew install GCC

PYTHON

Python is an interpreted, high-level, general-purpose programming language. It is used in OpenSees as an interpreter

in the OpenSeesPy version. In OpenSeesPy, Python version 3 is used.
Installing PYTHON via brew:

brew install python

MISC. NOTES

For the SUPERLU library. The file supermatrix.h throws an undefined error for the type int_t. It is actually defined
in the file slu_ddefs.h, but for some reason the compiler is not linking the two. Add the following line, copied from
slu_ddefs.h to supermatrix.h around line 17:

typedef int int_t; /* default */

1.3 Change Log

• Version ‘3.4.0.7‘<https://github.com/zhuminjie/OpenSeesPy/commit/169d1228baa23b5ea4f520fa3f6b6428dfb6d44e>_
– Update modalDamping command.
– Making rigid diaphragm work in X-Y plane for 2D models
– Making equalDOF use all DOFs if no DOFs are specified
– Reverted ViscousDamper and BilinearOilDamper

1.3. Change Log 29

OpenSeesPy Documentation, Release 3.5.1.3

• Version ‘3.4.0.6‘<>_ * Add new pyversion command * Add new Series3D material wrapper * Add new
ten-node tetrahedron element * Rename getFixedNodes, getFixedDOFs, getConstrainedNodes,
getConstrainedDOFs, getRetainedNodes, getRetainedDOFs. * Add new ASDConcrete3D
Plastic-Damage model
• Version ‘3.4.0.5<https://github.com/zhuminjie/OpenSeesPy/tree/v3.4.0.5>‘_ (11/29/2022)
– update to commit.
– Fix ElasticBilin.
– Update SimpleFractureMaterial.
– Update ASDShellQ4 element.
– Update LineMesh so that last element nodes not flipped.
– Update BoucWenMaterial with parallel.
– Fix ElasticMaterial tangent.
– ZeroLength not require yaxis input for 2D and 3D models.
– Add a, b parameter in ENTMaterial.
– Add material testing command.
– Add HystereticSmooth and Hysteretic Asym materials.
– Update ElasticTimoshenkoBeam2D and ElasticTimoshenkoBeam3D
– Update PM4Sand and PM4Silt materials.
– Add ConcreteMcftNonlinear5 and ConcreteMcftNonlinear7 materials
– Add Elastic Isotropic thermal material
– Add Plate fiber thermal material
– Add J2PlasticityTermal material
– Add PlateRebarThermal material
– Add PlateFromPlaneStressThermal material
– Add beamWithHinges element
– Update Parallel and Series materials
– Update ElasticTubeSection3d
– Add HSSSectionIntegration
– Add FRCC material
– Add ElasticBDShearSection2d
– Add OpenSeesInfo variable
– Allow strings for section codes in SectionAggregator
– Add GNG Material
– Add Orthotropic material wrapper
– Update PFEM with SSI only structure
– Add getRVParamTag, getRVValue, gradPerformanceFunction for reliability
– Add '-ele' and '-node' to rayleigh command

30 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

– Add ResponseSpectrumAnalysis command

– Add CoulombDamper material
– Add damping
– Add command getPatterns
– Add PFEM 3D contact element
– Add '-noFlush' option to analyze
– Add ConcentratedPlasticity and ConcentratedCurvature beam integrations
• Version 3.4.0.2 (7/20/2022)
– Update to commit abebbee.
– Update sectionForce and sectionDeformation commands.
– Update Tcl to Python converter.
– Add IGA to OpenSees.
– Update reliability commands.
– Add DuctileFracture material
– Update mass normalization for full gen lapack eigen solver
– Add ReeseStiffClayBelowWS, ReeseStiffClayAboveWS, VuggyLimestone, CementedSoil, WeakRock
backbone functions
– Add 2D version of rigidDiaphragm
– Add ShellNLDKGQ to mesh command
– Update performanceFunction command
– Add PythonEvaluator
– Update startPoint, gradientEvaluator, randomNumberGenerator, searchDirection, meritFunctionCheck,
stepSizeRule, rootFinding commands
– Add runFOSMAnalysis command
– Add findDesignPoints command
– Add runFORMAnalysis command
– Add BoucWenInfill material
– Add TDConcrete material
– Add MultiplierMaterial
– Add TzSimple, PySimple2, SAniSandMS to OpenSeesPy
– Add SPSW02 to OpenSeesPy
– Add RegularizedHinge to beam integration
– Add HystereticAsym and HystereticSmooth materials
– Add fixedNodes and fixedDOFs commands
– Add fixed/constrained/retained-Nodes/DOFs commands
– Add constrainedDOFs and retainedDOFs commands
– Add getConstraintMatrix command

1.3. Change Log 31

OpenSeesPy Documentation, Release 3.5.1.3

– Add timoshenkoBeamColumn element

– Update NDFiberSection command
– Update SFI_MVLEM_3D element
– Update NDFiberSectionWarping2d material
– Add MaterialBackbone
– Update GradientInelasticBeamColumn3d
– Update ElasticBeam2d and 3d
• Version 3.4.0.1 (12/22/2021)
– Update to OpenSees version 3.4.0 at commit 3b28d0c.
– Add command for QzSimple2 material
– Updates to PFEM
– Add masonry elements
– Add DBNL3D
– Update to MixedBeamColumn
– Add SMA material
– Update to material test
– Update to IMKBilin material
– Update to J2CyclicBoundingSurface model
– Add InertiaTruss element
– Add SteelFractureDI material
– Update to SectionIngegration
– Update to HystereticMaterial
– Add PySimple2 and TzSimple2 materials
– Update to MultiLinear material
– Add eleType command
– Add ExpressNewton command
– Add sectionTag command
– Update to ASDAbsorbingBoundary3D element
– Add TDConcrete elements
– Update to forceBeamColumn
– Add HSS to section command
– Add SteelDRC material
– Add SAniSandMS material
– Add getNodeLoadTags command
– Add getNDF command
– Add startPoint command

32 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

– Add getCrdTransfTags command

– Add randomNumberGenerator command
– Add commands for step size rule, function evaluator, and root finding
– Add gradientEvaluator and performanceFunction commands
– Add HarmonicSteadyState integrator
– Add DowelType material
– Add double membrane plate fiber section
• Version 3.3.0.0 (6/4/2021)
– Update to 3.3.0.0 at commit 5c925e6.
• Version 3.2.2.9 (1/28/2021)
– The pip installation will only install needed libraries based on the Operating System, i.e. Windows, Linux,
or Mac. The installation time and size are now one third of before.
– Bug fixes for Pinching4Material, Concrete07, H5DRM, RCCircularSectionIntegration, ResponseSpectru-
mAnalysis, PML, FiberSection2d, DriftRecorder,
• Version 3.2.2.8 (1/8/2021)
– Linux version is tested with Centos 7, 8, Ubuntu 18.04, 20.04, Fedora, and Debian.
– Mac version uses MacPorts for installing Python and dependencies.
– Bug fixes for recorders, FourNodeTetrahedron, ASD_SMA_3K, nodeMass,
– Add ExpressNewton, RockingBC, CBDI3d, Concrete02IS
– Update PETSc Solver, ZeroLengthSection, ForceBeamColumn3d, OOHystereticMaterial, SSPbrickUP,
HardeningMaterial, BilinearOilDamper,
• Version 3.2.2.6 (10/15/2020)
– OpenSeesPy is available now on Mac, just type import openseespy.opensees as ops on the MacOS.
Python3.8 is required and HomeBrew Python is strongly recommended.
– SixNodeTri element by Seweryn
– PostProcessing package ops_vis by Seweryn
• Version 3.2.2.5 (9/16/2020)
– Fix a Windows issue for virtual environment
• Version 3.2.2.4 (9/10/2020)
– Bug fixes in Truss and ForceBeamColumn2d
– Adding ops.__version__ variable
– Bug fixes in 3D elastic beam
– Bug fixes in Tri31
– Adding ops_vis module for plotting
– OpenSees commit b0f6b06
• Version 3.2.2.3 (8/11/2020)
– Fix typos in documentation

1.3. Change Log 33

OpenSeesPy Documentation, Release 3.5.1.3

– Add testNorm and testIter commands

– Add Python3.7 and Python 3.8 support
– Support latest Anaconda
– Improvements of ploting commands
– ShellDKGT command
– Include OpenSees commits upto 380239c on 8/9
• Version 3.2.2.1 (5/18/2020)
– add gimmeMCK integrator
• Version 3.2.2 (5/8/2020)
– Fix Get_Renderiing tab problem
– Ship with dependent libraries for more Linux systems
• Version 3.2.0 (4.17.2020)
– Add background mesh command
– Add partition command
– Add OpenSeesPy test
– Many bug fixes
• Version 3.1.5.11 (1.10.2020)
– Change versioning method. First two digits match the current OpenSees framework version. The last two
digits are the versions for OpenSeesPy.
– For Windows, only support the Python version that corresponds to the current version of Anaconda.
– Add openseespy.postprocessing.Get_Rendering
– Add ‘-init’ option to Newmark integrator
– Some function can return empty or one-element lists
– Spaces in string input will be automatically removed
– Bug fixes
• Version 0.5.4
– Support Mac
– Support Python3.7 on Windows and Linux
• Version 0.5.3
– Fix bug in LimitState UniaxialMaterial
– Automatic trimming spaces for string inputs
– Some output commands return lists instead of ints, such as nodeDisp etc.
• Version 0.5.2
– Add package openseespy.postprocessing
– Add setStartNodeTag command
– modalDamping: bug fixes

34 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

– Add Steel02Fatiuge material

– Add Concrete02IS material
– Add HardeningMaterial2 material
– Add hystereticBackone command
– Add stiffnessDegradation command
– Add strengthDegradation command
– Add unloadingRule command
• Version 0.4.2019.7
– Parallel: the Linux version is enabled with parallel capability
– Python stream: add no echo
– Mesh: add CorotTruss
– TriMesh: can create line elements
– QuadMesh: can create line and triangular elements
– Python inputs: more flexible input types
– Commands: add ExplicitDifference integrator
• Version 0.3.0
– Add logFile command
– Add partial uniform load fo ForceBeamColumn
– Add ShellDKGT element
– Add ‘-V’ option in Newmark and HHT
– Fix bugs in wipe and Mesh
– Various PFEM updates
– Update to OpenSees 3.0.3
• Version 0.2.0 (8a3d622)
– OpenSeesPy now can print messages and errors in Jupyter Notebook and other Windows based Python
applications
– Add setParameter command
– Add nodeDOFs command
– Add setNumThread and getNumThread commands in a multi-threaded environment
– Add logFile command
– printA and prinbB can return matrix and vector as lists
– Fix bugs in updateMaterialStage
– PM4Sand improvements
– Add CatenaryCable element to OpenSeesPy
• Version 0.1.1 (f9f45fe)
– Update to OpenSees 3.0.2

1.3. Change Log 35

OpenSeesPy Documentation, Release 3.5.1.3

• Version 0.0.7 (b75db21)

– Add “2D wheel-rail” element
– PVD recorder allows to set a path
– Add “sdfResponse” function for single dof dynamic analysis
– Fix a bug in Joint2D
– Fix typo in UCSD UP elements
– Fix bugs in PressureIndependMultiYield
– Add JSON print options to some materials and elements
• Version 0.0.6 (cead6e8)
– Add “nonlinearBeamColumn” element for backward compatability
– Add “updateMaterialStage” function
– Add “RCCircular” seciton
– Add “quadr” patch for backward compatibility
– Fix bugs in “Steel01Thermal” material
– Fix bugs in Truss
– Fix bugs in eleNodes function
– Fix bugs in ZeroLength element
– Fix bugs in FiberSection2d
– Fix bugs in PFEMLinSOE
• Version 0.0.5 (215c63d)
– Update to OpenSees 3.0.0

1.4 Model Commands

The model or domain in OpenSees is a collection (an aggregation in object-oriented terms) of elements, nodes, single-
and multi-point constraints and load patterns. It is the aggregation of these components which define the type of model
that is being analyzed.
1. model command
2. element commands
3. node command
4. sp constraint commands
5. mp constraint commands
6. pressureConstraint command
7. timeSeries commands
8. pattern commands
9. mass command
10. region command

36 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

11. rayleigh command

12. block commands
13. beamIntegration commands
14. uniaxialMaterial commands
15. nDMaterial commands
16. section commands
17. frictionModel commands
18. geomTransf commands

1.4.1 model command

model(’basic’, ’-ndm’, ndm, ’-ndf’, ndf=ndm*(ndm+1)/2)

Set the default model dimensions and number of dofs.

ndm (int) number of dimensions (1,2,3)

ndf (int) number of dofs (optional)

1.4.2 element commands

element(eleType, eleTag, *eleNodes, *eleArgs)

Create a OpenSees element.

eleType (str) element type

eleTag (int) element tag.
eleNodes (list (int)) a list of element nodes, must be preceded with *.
eleArgs (list) a list of element arguments, must be preceded with *.

For example,

eleType = 'truss'
eleTag = 1
eleNodes = [iNode, jNode]
eleArgs = [A, matTag]
element(eleType, eleTag, *eleNodes, *eleArgs)

The following contain information about available eleType:

Zero-Length Element

1. zeroLength Element
2. zeroLengthND Element
3. zeroLengthSection Element
4. CoupledZeroLength Element
5. zeroLengthContact Element
6. zeroLengthContactNTS2D

1.4. Model Commands 37

OpenSeesPy Documentation, Release 3.5.1.3

7. zeroLengthInterface2D
8. zeroLengthImpact3D

zeroLength Element

element(’zeroLength’, eleTag, *eleNodes, ’-mat’, *matTags, ’-dir’, *dirs, <’-doRayleigh’, rFlag=0>, <’-
orient’, *vecx, *vecyp>)
This command is used to construct a zeroLength element object, which is defined by two nodes at the same
location. The nodes are connected by multiple UniaxialMaterial objects to represent the force-deformation
relationship for the element.

eleTag (int) unique element object tag

eleNodes (list (int)) a list of two element nodes
matTags (list (int)) a list of tags associated with previously-defined Uni-
axialMaterials
dirs (list (int)) a list of material directions:
• 1,2,3 - translation along local x,y,z axes, respectively;

– 4,5,6 - rotation about local x,y,z axes,

respectively

rFlag (float) optional, default = 0 NO RAYLEIGH DAMPING

(default), 1 include Rayleigh damping
vecx (list (float)) a list of vector components in global coordinates
defining local x-axis (optional)
vecyp (list (float)) a list of vector components in global coordinates
defining vector yp which lies in the local x-y plane
for the element. (optional)

Note: If the optional orientation vectors are not specified, the local element axes coincide with the global axes.
Otherwise the local z-axis is defined by the cross product between the vectors x and yp vectors specified on the
command line.

See also:
Notes

zeroLengthND Element

element(’zeroLengthND’, eleTag, *eleNodes, matTag, <uniTag>, <’-orient’, *vecx, vecyp>)

This command is used to construct a zeroLengthND element object, which is defined by two nodes at the same
location. The nodes are connected by a single NDMaterial object to represent the force-deformation relationship
for the element.

38 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

eleTag unique element object tag

(int)
eleNodes a list of two element nodes
(list (int))
matTag tag associated with previously-defined ndMaterial object
(int)
uniTag tag associated with previously-defined UniaxialMaterial object which may be used to represent
(int) uncoupled behavior orthogonal to the plane of the NDmaterial response. SEE NOTES 2 and
3.
vecx (list a list of vector components in global coordinates defining local x-axis (optional)
(float))
vecyp a list of vector components in global coordinates defining vector yp which lies in the local x-y
(list plane for the element. (optional)
(float))

Note:
1. The zeroLengthND element only represents translational response between its nodes
2. If the NDMaterial object is of order two, the response lies in the element local x-y plane and the UniaxialMaterial
object may be used to represent the uncoupled behavior orthogonal to this plane, i.e. along the local z-axis.
3. If the NDMaterial object is of order three, the response is along each of the element local exes.
4. If the optional orientation vectors are not specified, the local element axes coincide with the global axes. Oth-
erwise the local z-axis is defined by the cross product between the vectors x and yp vectors specified on the
command line.
5. The valid queries to a zero-length element when creating an ElementRecorder object are ‘force’, ‘deformation’,
and ‘material matArg1 matArg2 . . . ’

See also:
Notes

zeroLengthSection Element

element(’zeroLengthSection’, eleTag, *eleNodes, secTag, <’-orient’, *vecx, *vecyp>, <’-doRayleigh’,

rFlag>)
This command is used to construct a zero length element object, which is defined by two nodes at the same
location. The nodes are connected by a single section object to represent the force-deformation relationship for
the element.

1.4. Model Commands 39

OpenSeesPy Documentation, Release 3.5.1.3

eleTag (int) unique element object tag

eleNodes (list (int)) a list of two element nodes
secTag (int) tag associated with previously-defined Section ob-
ject
vecx (list (float)) a list of vector components in global coordinates
defining local x-axis (optional)
vecyp (list (float)) a list of vector components in global coordinates
defining vector yp which lies in the local x-y plane
for the element. (optional)
rFlag (float) optional, default = 0
• rFlag = 0 NO RAYLEIGH DAMPING (de-
fault)
• rFlag = 1 include rayleigh damping

See also:
Notes

CoupledZeroLength Element

element(’CoupledZeroLength’, eleTag, *eleNodes, dirn1, dirn2, matTag, <rFlag=1>)

eleTag (int) unique element object tag

eleNodes (list (int)) a list of two element nodes
matTag (float) tags associated with previously-defined UniaxialMa-
terial
dirn1 dirn2 (int) the two directions, 1 through ndof.
rFlag (float) optional, default = 0
• rFlag = 0 NO RAYLEIGH DAMPING (de-
fault)
• rFlag = 1 include rayleigh damping

See also:
Notes

zeroLengthContact Element

element(’zeroLengthContact2D’, eleTag, *eleNodes, Kn, Kt, mu, ’-normal’, Nx, Ny)

This command is used to construct a zeroLengthContact2D element, which is Node-to-node frictional contact
element used in two dimensional analysis and three dimensional analysis:

eleTag (int) unique element object tag

eleNodes (list (int)) a list of a constrained and a retained nodes
Kn (float) Penalty in normal direction
Kt (float) Penalty in tangential direction
mu (float) friction coefficient

40 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

element(’zeroLengthContact3D’, eleTag, *eleNodes, Kn, Kt, mu, c, dir)

This command is used to construct a zeroLengthContact3D element, which is Node-to-node frictional contact
element used in two dimensional analysis and three dimensional analysis:

eleTag (int) unique element object tag

eleNodes (list (int)) a list of a constrained and a retained nodes
Kn (float) Penalty in normal direction
Kt (float) Penalty in tangential direction
mu (float) friction coefficient
c (float) cohesion (not available in 2D)
dir (int) Direction flag of the contact plane (3D), it can be:
• 1 Out normal of the master plane pointing to
+X direction
• 2 Out normal of the master plane pointing to
+Y direction
• 3 Out normal of the master plane pointing to
+Z direction

See also:
Notes

zeroLengthContactNTS2D

element(’zeroLengthContactNTS2D’, eleTag, ’-sNdNum’, sNdNum, ’-mNdNum’, mNdNum, ’-Nodes’,

*NodesTags, kn, kt, phi)

eleTag (int) unique element object tag

sNdNum (int) Number of Slave Nodes
mNdNum (int) Number of Master nodes
NodesTags (list (int)) Slave and master node tags respectively
kn (float) Penalty in normal direction
kt (float) Penalty in tangential direction
phi (float) Friction angle in degrees

Note:
1. The contact element is node-to-segment (NTS) contact. The relation follows Mohr-Coulomb frictional law:
𝑇 = 𝑁 × tan(𝜑), where 𝑇 is the tangential force, 𝑁 is normal force across the interface and 𝜑 is friction angle.
2. For 2D contact, slave nodes and master nodes must be 2 DOF and notice that the slave and master nodes must
be entered in counterclockwise order.
3. The resulting tangent from the contact element is non-symmetric. Switch to the non-symmetric matrix solver if
convergence problem is experienced.
4. As opposed to node-to-node contact, predefined normal vector for node-to-segment (NTS) element is not re-
quired because contact normal will be calculated automatically at each step.
5. contact element is implemented to handle large deformations.

See also:

1.4. Model Commands 41

OpenSeesPy Documentation, Release 3.5.1.3

Notes

zeroLengthInterface2D

element(’zeroLengthInterface2D’, eleTag, ’-sNdNum’, sNdNum, ’-mNdNum’, mNdNum, ’-dof’, sdof, mdof,

’-Nodes’, *NodesTags, kn, kt, phi)

eleTag (int) unique element object tag

sNdNum (int) Number of Slave Nodes
mNdNum (int) Number of Master nodes
sdof, mdof (int) Slave and Master degree of freedom
NodesTags (list (int)) Slave and master node tags respectively
kn (float) Penalty in normal direction
kt (float) Penalty in tangential direction
phi (float) Friction angle in degrees

Note:
1. The contact element is node-to-segment (NTS) contact. The relation follows Mohr-Coulomb frictional law:
𝑇 = 𝑁 × tan(𝜑), where 𝑇 is the tangential force, 𝑁 is normal force across the interface and 𝜑 is friction angle.
2. For 2D contact, slave nodes and master nodes must be 2 DOF and notice that the slave and master nodes must
be entered in counterclockwise order.
3. The resulting tangent from the contact element is non-symmetric. Switch to the non-symmetric matrix solver if
convergence problem is experienced.
4. As opposed to node-to-node contact, predefined normal vector for node-to-segment (NTS) element is not re-
quired because contact normal will be calculated automatically at each step.
5. contact element is implemented to handle large deformations.

See also:
Notes

zeroLengthImpact3D

element(’zeroLengthImpact3D’, eleTag, *eleNodes, direction, initGap, frictionRatio, Kt, Kn, Kn2, Delta_y,
cohesion)
This command constructs a node-to-node zero-length contact element in 3D space to simulate the im-
pact/pounding and friction phenomena.

42 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

eleTag (int) unique element object tag

eleNodes (list (int)) a list of a constrained and a retained nodes
direction (int)
• 1 if out-normal vector of master plane points to
+X direction
• 2 if out-normal vector of master plane points to
+Y direction
• 3 if out-normal vector of master plane points to
+Z direction

initGap (float) Initial gap between master plane and slave plane
frictionRatio (float) Friction ratio in two tangential directions (parallel to
master and slave planes)
Kt (float) Penalty in two tangential directions
Kn (float) Penalty in normal direction (normal to master and
slave planes)
Kn2 (float) Penalty in normal direction after yielding based on
Hertz impact model
Delta_y (float) Yield deformation based on Hertz impact model
cohesion (float) Cohesion, if no cohesion, it is zero

Note:
1. This element has been developed on top of the “zeroLengthContact3D”. All the notes available in “zeroLength-
Contact3D” wiki page would apply to this element as well. It includes the definition of master and slave nodes,
the number of degrees of freedom in the domain, etc.
2. Regarding the number of degrees of freedom (DOF), the end nodes of this element should be defined in 3DOF
domain. For getting information on how to use 3DOF and 6DOF domain together, please refer to OpenSees
documentation and forums or see the zip file provided in the EXAMPLES section below.
3. This element adds the capabilities of “ImpactMaterial” to “zeroLengthContact3D.”
4. For simulating a surface-to-surface contact, the element can be defined for connecting the nodes on slave surface
to the nodes on master surface.
5. The element was found to be fast-converging and eliminating the need for extra elements and nodes in the
modeling process.

See also:
Notes

Truss Elements

1. Truss Element
2. Corotational Truss Element

Truss Element

This command is used to construct a truss element object. There are two ways to construct a truss element object:

1.4. Model Commands 43

OpenSeesPy Documentation, Release 3.5.1.3

element(’Truss’, eleTag, *eleNodes, A, matTag, <’-rho’, rho>, <’-cMass’, cFlag>, <’-doRayleigh’,

rFlag>)
One way is to specify an area and a UniaxialMaterial identifier:
element(’TrussSection’, eleTag, *eleNodes, secTag, <’-rho’, rho>, <’-cMass’, cFlag>, <’-doRayleigh’,
rFlag>)
the other is to specify a Section identifier:

eleTag (int) unique element object tag

eleNodes (list (int)) a list of two element nodes
A (float) cross-sectional area of element
matTag (int) tag associated with previously-defined UniaxialMa-
terial
secTag (int) tag associated with previously-defined Section
rho (float) mass per unit length, optional, default = 0.0
cFlag (float) consistent mass flag, optional, default = 0
• cFlag = 0 lumped mass matrix (default)
• cFlag = 1 consistent mass matrix

rFlag (float) Rayleigh damping flag, optional, default = 0

• rFlag = 0 NO RAYLEIGH DAMPING (de-
fault)
• rFlag = 1 include Rayleigh damping

Note:
1. The truss element DOES NOT include geometric nonlinearities, even when used with beam-columns utilizing
P-Delta or Corotational transformations.
2. When constructed with a UniaxialMaterial object, the truss element considers strain-rate effects, and is thus
suitable for use as a damping element.
3. The valid queries to a truss element when creating an ElementRecorder object are ‘axialForce,’ ‘forces,’ ‘lo-
calForce’, deformations,’ ‘material matArg1 matArg2. . . ,’ ‘section sectArg1 sectArg2. . . ’ There will be more
queries after the interface for the methods involved have been developed further.

See also:
Notes

Corotational Truss Element

This command is used to construct a corotational truss element object. There are two ways to construct a corotational
truss element object:
element(’corotTruss’, eleTag, *eleNodes, A, matTag, <’-rho’, rho>, <’-cMass’, cFlag>, <’-doRayleigh’,
rFlag>)
One way is to specify an area and a UniaxialMaterial identifier:
element(’corotTrussSection’, eleTag, *eleNodes, secTag, <’-rho’, rho>, <’-cMass’, cFlag>, <’-
doRayleigh’, rFlag>)
the other is to specify a Section identifier:

44 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

eleTag (int) unique element object tag

eleNodes (list (int)) a list of two element nodes
A (float) cross-sectional area of element
matTag (int) tag associated with previously-defined UniaxialMa-
terial
secTag (int) tag associated with previously-defined Section
rho (float) mass per unit length, optional, default = 0.0
cFlag (float) consistent mass flag, optional, default = 0
• cFlag = 0 lumped mass matrix (default)
• cFlag = 1 consistent mass matrix

rFlag (float) Rayleigh damping flag, optional, default = 0

• rFlag = 0 NO RAYLEIGH DAMPING (de-
fault)
• rFlag = 1 include Rayleigh damping

Note:
1. When constructed with a UniaxialMaterial object, the corotational truss element considers strain-rate effects,
and is thus suitable for use as a damping element.
2. The valid queries to a truss element when creating an ElementRecorder object are ‘axialForce,’ ‘stiff,’ defor-
mations,’ ‘material matArg1 matArg2. . . ,’ ‘section sectArg1 sectArg2. . . ’ There will be more queries after the
interface for the methods involved have been developed further.
3. CorotTruss DOES NOT include Rayleigh damping by default.

See also:
Notes

Beam-Column Elements

1. Elastic Beam Column Element

2. Elastic Beam Column Element with Stiffness Modifiers
3. Elastic Timoshenko Beam Column Element
4. Beam With Hinges Element
5. dispBeamColumn
6. forceBeamColumn
7. nonlinearBeamColumn
8. Flexure-Shear Interaction Displacement-Based Beam-Column Element
9. MVLEM - Multiple-Vertical-Line-Element-Model for RC Walls
10. SFI MVLEM - Cyclic Shear-Flexure Interaction Model for RC Walls

1.4. Model Commands 45

OpenSeesPy Documentation, Release 3.5.1.3

Elastic Beam Column Element

This command is used to construct an elasticBeamColumn element object. The arguments for the construction of an
elastic beam-column element depend on the dimension of the problem, (ndm)
element(’elasticBeamColumn’, eleTag, *eleNodes, Area, E_mod, Iz, transfTag, <’-mass’, mass>, <’-
cMass’>, <’-release’, releaseCode>)
element(’elasticBeamColumn’, eleTag, *eleNodes, secTag, transfTag, <’-mass’, mass>, <’-cMass’>, <’-
release’, releaseCode>)
For a two-dimensional problem
element(’elasticBeamColumn’, eleTag, *eleNodes, Area, E_mod, G_mod, Jxx, Iy, Iz, transfTag, <’-mass’,
mass>, <’-cMass’>)
element(’elasticBeamColumn’, eleTag, *eleNodes, secTag, transfTag, <’-mass’, mass>, <’-cMass’> <’-
releasez’, releaseCode>, <’-releasey’, releaseCode>)
For a three-dimensional problem

eleTag (int) unique element object tag

eleNodes (list a list of two element nodes
(int))
Area (float) cross-sectional area of element
E_mod (float) Young’s Modulus
G_mod (float) Shear Modulus
Jxx (float) torsional moment of inertia of cross section
Iy (float) second moment of area about the local y-axis
Iz (float) second moment of area about the local z-axis
secTag (int) identifier for previously-defined section object
transfTag (int) identifier for previously-defined coordinate-transformation (CrdTransf) object
mass (float) element mass per unit length (optional, default = 0.0)
'-cMass' (str) to form consistent mass matrix (optional, default = lumped mass matrix)
'releaseCode' moment release (optional, 0=no release (default), 1=release at I, 2=release at J, 3=re-
(int) lease at I and J)

See also:
Notes

Elastic Beam Column Element with Stiffness Modifiers

This command is used to construct a ModElasticBeam2d element object. The arguments for the construction of an
elastic beam-column element with stiffness modifiers is applicable for 2-D problems. This element should be used for
modelling of a structural element with an equivalent combination of one elastic element with stiffness-proportional
damping, and two springs at its two ends with no stiffness proportional damping to represent a prismatic section. The
modelling technique is based on a number of analytical studies discussed in Zareian and Medina (2010) and Zareian
and Krawinkler (2009) and is utilized in order to solve problems related to numerical damping in dynamic analysis of
frame structures with concentrated plasticity springs.
element(’ModElasticBeam2d’, eleTag, *eleNodes, Area, E_mod, Iz, K11, K33, K44, transfTag, <’-mass’,
massDens>, <’-cMass’>)

46 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

eleTag (int) unique element object tag

eleNodes (list (int)) a list of two element nodes
Area (float) cross-sectional area of element
E_mod (float) Young’s Modulus
Iz (float) second moment of area about the local z-axis
K11 (float) stiffness modifier for translation
K33 (float) stiffness modifier for translation
K44 (float) stiffness modifier for rotation
transfTag (int) identifier for previously-defined coordinate-transformation (CrdTransf) object
massDens (float) element mass per unit length (optional, default = 0.0)
'-cMass' (str) to form consistent mass matrix (optional, default = lumped mass matrix)

See also:
Notes

Elastic Timoshenko Beam Column Element

This command is used to construct an ElasticTimoshenkoBeam element object. A Timoshenko beam is a frame
member that accounts for shear deformations. The arguments for the construction of an elastic Timoshenko beam
element depend on the dimension of the problem, ndm:
element(’ElasticTimoshenkoBeam’, eleTag, *eleNodes, E_mod, G_mod, Area, Iz, Avy, transfTag, <’-mass’,
massDens>, <’-cMass’>)
For a two-dimensional problem:
element(’ElasticTimoshenkoBeam’, eleTag, *eleNodes, E_mod, G_mod, Area, Iz, Jxx, Iy, Iz, Avy, Avz, trans-
fTag, <’-mass’, massDens>, <’-cMass’>)
For a three-dimensional problem:

eleTag (int) unique element object tag

eleNodes (list (int)) a list of two element nodes
E_mod (float) Young’s Modulus
G_mod (float) Shear Modulus
Area (float) cross-sectional area of element
Jxx (float) torsional moment of inertia of cross section
Iy (float) second moment of area about the local y-axis
Iz (float) second moment of area about the local z-axis
Avy (float) Shear area for the local y-axis
Avz (float) Shear area for the local z-axis
transfTag (int) identifier for previously-defined coordinate-transformation (CrdTransf) object
massDens (float) element mass per unit length (optional, default = 0.0)
'-cMass' (str) to form consistent mass matrix (optional, default = lumped mass matrix)

See also:
Notes

Beam With Hinges Element

This command is used to construct a forceBeamColumn element object, which is based on the non-iterative (or it-
erative) flexibility formulation. The locations and weights of the element integration points are based on so-called

1.4. Model Commands 47

OpenSeesPy Documentation, Release 3.5.1.3

plastic hinge integration, which allows the user to specify plastic hinge lenghts at the element ends. Two-point Gauss
integration is used on the element interior while two-point Gauss-Radau integration is applied over lengths of 4LpI
and 4LpJ at the element ends, viz. “modified Gauss-Radau plastic hinge integration”. A total of six integration points
are used in the element state determination (two for each hinge and two for the interior).
Users may be familiar with the beamWithHinges command format (see below); however, the format shown here allows
for the simple but important case of using a material nonlinear section model on the element interior. The previous
beamWithHinges command constrained the user to an elastic interior, which often led to unconservative estimates of
the element resisting force when plasticity spread beyond the plastic hinge regions in to the element interior.
The advantages of this new format over the previous beamWithHinges command are
• Plasticity can spread beyond the plastic hinge regions
• Hinges can form on the element interior, e.g., due to distributed member loads
To create a beam element with hinges, one has to use a forceBeamColumn element with following
beamIntegration().

Note:
• 'HingeRadau' – two-point Gauss-Radau applied to the hinge regions over 4LpI and 4LpJ (six element inte-
gration points)
• 'HingeRadauTwo' – two-point Gauss-Radau in the hinge regions applied over LpI and LpJ (six element
integration points)
• 'HingeMidpoint' – midpoint integration over the hinge regions (four element integration points)
• 'HingeEndpoint' – endpoint integration over the hinge regions (four element integration points)

See also:
For more information on the behavior, advantages, and disadvantages of these approaches to plastic hinge integration,
see
Scott, M.H. and G.L. Fenves. “Plastic Hinge Integration Methods for Force-Based Beam-Column Elements”, Journal
of Structural Engineering, 132(2):244-252, February 2006.
Scott, M.H. and K.L. Ryan. “Moment-Rotation Behavior of Force-Based Plastic Hinge Elements”, Earthquake Spec-
tra, 29(2):597-607, May 2013.
The primary advantages of HingeRadau are
• The user can specify a physically meaningful plastic hinge length
• The largest bending moment is captured at the element ends
• The exact numerical solution is recovered for a linear-elastic prismatic beam
• The characteristic length is equal to the user-specified plastic hinge length when deformations localize at the
element ends
while the primary disadvantages are
• The element post-yield response is too flexible for strain-hardening section response (consider using HingeR-
adauTwo)
• The user needs to know the plastic hinge length a priori (empirical equations are available)

48 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

dispBeamColumn

element(’dispBeamColumn’, eleTag, *eleNodes, transfTag, integrationTag, ’-cMass’, ’-mass’, mass=0.0)

Create a dispBeamColumn element.

eleTag (int) tag of the element

eleNodes (list (int)) list of two node tags
transfTag (int) tag of transformation
integrationTag tag of beamIntegration()
(int)
'-cMass' to form consistent mass matrix (optional, default = lumped mass matrix)
mass (float) element mass density (per unit length), from which a lumped-mass matrix is
formed (optional)

forceBeamColumn

element(’forceBeamColumn’, eleTag, *eleNodes, transfTag, integrationTag, ’-iter’, maxIter=10, tol=1e-12,

’-mass’, mass=0.0)
Create a ForceBeamColumn element.

eleTag (int) tag of the element

eleNodes (list (int)) a list of two element nodes
transfTag (int) tag of transformation
integrationTag tag of beamIntegration()
(int)
maxIter (int) maximum number of iterations to undertake to satisfy element compatibility (op-
tional)
tol (float) tolerance for satisfaction of element compatibility (optional)
mass (float) element mass density (per unit length), from which a lumped-mass matrix is
formed (optional)

nonlinearBeamColumn

element(’nonlinearBeamColumn’, eleTag, *eleNodes, numIntgrPts, secTag, transfTag, ’-iter’, maxIter=10,

tol=1e-12, ’-mass’, mass=0.0, ’-integration’, intType)
Create a nonlinearBeamColumn element. This element is for backward compatability.

1.4. Model Commands 49

OpenSeesPy Documentation, Release 3.5.1.3

eleTag (int) tag of the element

eleNodes (list (int)) a list of two element nodes
numIntgrPts (int) number of integration points.
secTag (int) tag of section
transfTag (int) tag of transformation
maxIter (int) maximum number of iterations to undertake to sat-
isfy element compatibility (optional)
tol (float) tolerance for satisfaction of element compatibility
(optional)
mass (float) element mass density (per unit length), from which a
lumped-mass matrix is formed (optional)
intType (str) integration type (optional, default is 'Lobatto')
• 'Lobatto'
• 'Legendre'
• 'Radau'
• 'NewtonCotes'
• 'Trapezoidal'

Flexure-Shear Interaction Displacement-Based Beam-Column Element

This command is used to construct a dispBeamColumnInt element object, which is a distributed-plasticity,

displacement-based beam-column element which includes interaction between flexural and shear components.
element(’dispBeamColumnInt’, eleTag, *eleNodes, numIntgrPts, secTag, transfTag, cRot, <’-mass’, mass-
Dens>)

eleTag unique element object tag

(int)
eleNodes a list of two element nodes
(list (int))
numIntgrPtsnumber of integration points along the element.
(int)
secTag identifier for previously-defined section object
(int)
transfTag identifier for previously-defined coordinate-transformation (CrdTransf) object
(int)
cRot (float) identifier for element center of rotation (or center of curvature distribution). Fraction of the
height distance from bottom to the center of rotation (0 to 1)
massDens element mass density (per unit length), from which a lumped-mass matrix is formed (op-
(float) tional, default=0.0)

See also:
Notes

MVLEM - Multiple-Vertical-Line-Element-Model for RC Walls

Developed and implemented by:

Kristijan Kolozvari (CSU Fullerton)
Kutay Orakcal (Bogazici University)

50 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

John Wallace (UCLA)

The MVLEM element command is used to generate a two-dimensional Multiple-Vertical-Line-Element-Model

(MVLEM; Vulcano et al., 1988; Orakcal et al., 2004, Kolozvari et al., 2015) for simulation of flexure-dominated
RC wall behavior. A single model element incorporates six global degrees of freedom, three of each located at the
center of rigid top and bottom beams, as illustrated in Figure 1a. The axial/flexural response of the MVLEM is sim-
ulated by a series of uniaxial elements (or macro-fibers) connected to the rigid beams at the top and bottom (e.g.,
floor) levels, whereas the shear response is described by a shear spring located at height ch from the bottom of the
wall element (Figure 1a). Shear and flexural responses of the model element are uncoupled. The relative rotation
between top and bottom faces of the wall element occurs about the point located on the central axis of the element at
height ch (Figure 1b). Rotations and resulting transverse displacements are calculated based on the wall curvature, de-
rived from section and material properties, corresponding to the bending moment at height ch of each element (Figure
1b). A value of c=0.4 was recommended by Vulcano et al. (1988) based on comparison of the model response with
experimental results.
element(’MVLEM’, eleTag, Dens, *eleNodes, m, c, ’-thick’, *thick, ’-width’, *widths, ’-rho’, *rho, ’-
matConcrete’, *matConcreteTags, ’-matSteel’, *matSteelTags, ’-matShear’, matShearTag)

eleTag (int) unique element object tag

Dens (float) Wall density
eleNodes (list (int)) a list of two element nodes
m (int) Number of element macro-fibers
c (float) Location of center of rotation from the iNode, c = 0.4 (recommended)
thick (list (float)) a list of m macro-fiber thicknesses
widths (list (float)) a list of m macro-fiber widths
rho (list (float)) a list of m reinforcing ratios corresponding to macro-fibers; for each fiber: 𝑟ℎ𝑜𝑖 =
𝐴𝑠,𝑖 /𝐴𝑔𝑟𝑜𝑠𝑠,𝑖 (1 < 𝑖 < 𝑚)
matConcreteTags a list of m uniaxialMaterial tags for concrete
(list (int))
matSteelTags (list a list of m uniaxialMaterial tags for steel
(int))
matShearTag (int) Tag of uniaxialMaterial for shear material

See also:
Notes
Kolozvari K., Orakcal K., and Wallace J. W. (2015a). “New opensees models for simulating nonlin-
ear flexural and coupled shear-flexural behavior of RC walls and columns”, Computers and Structures,
Volume 196, February 2018, Pages 246-262, doi

SFI MVLEM - Cyclic Shear-Flexure Interaction Model for RC Walls

Developed and implemented by:

Kristijan Kolozvari (CSU Fullerton)
Kutay Orakcal (Bogazici University)
Leonardo Massone (University of Chile, Santiago)
John Wallace (UCLA)

The SFI_MVLEM command is used to construct a Shear-Flexure Interaction Multiple-Vertical-Line-Element Model

(SFI-MVLEM, Kolozvari et al., 2018, 2015a, b, c; Kolozvari 2013), which captures interaction between axial/flexural

1.4. Model Commands 51

OpenSeesPy Documentation, Release 3.5.1.3

and shear behavior of RC structural walls and columns under cyclic loading. The SFI_MVLEM element (Figure 1)
incorporates 2-D RC panel behavior described by the Fixed-Strut-Angle-Model (nDMaterial FSAM; Ulugtekin, 2010;
Orakcal et al., 2012), into a 2-D macroscopic fiber-based model (MVLEM). The interaction between axial and shear
behavior is captured at each RC panel (macro-fiber) level, which further incorporates interaction between shear and
flexural behavior at the SFI_MVLEM element level.
element(’SFI_MVLEM’, eleTag, *eleNodes, m, c, ’-thick’, *thick, ’-width’, *widths, ’-mat’, *mat_tags)

eleTag (int) unique element object tag

eleNodes (list (int)) a list of two element nodes
m (int) Number of element macro-fibers
c (float) Location of center of rotation with from the iNode, c = 0.4 (recommended)
Thicknesses (list (float)) a list of m macro-fiber thicknesses
Widths (list (float)) a list of m macro-fiber widths
Material_tags (list (int)) a list of m macro-fiber nDMaterial1 tags

See also:
Notes
Kolozvari K., Orakcal K., and Wallace J. W. (2015a). “New opensees models for simulating nonlin-
ear flexural and coupled shear-flexural behavior of RC walls and columns”, Computers and Structures,
Volume 196, February 2018, Pages 246-262, doi
Kolozvari K., Orakcal K., and Wallace J. W. (2015a). ”Modeling of Cyclic Shear-Flexure Interaction
in Reinforced Concrete Structural Walls. I: Theory”, ASCE Journal of Structural Engineering, 141(5),
04014135 doi
Kolozvari K., Tran T., Orakcal K., and Wallace, J.W. (2015c). ”Modeling of Cyclic Shear-Flexure Inter-
action in Reinforced Concrete Structural Walls. II: Experimental Validation”, ASCE Journal of Structural
Engineering, 141(5), 04014136 doi
Kolozvari K., Orakcal K., and Wallace J. W. (2015c). “Shear-Flexure Interaction Modeling of reinforced
Concrete Structural Walls and Columns under Reversed Cyclic Loading”, Pacific Earthquake Engineering
Research Center, University of California, Berkeley, PEER Report No. 2015/12
Kolozvari K. (2013). “Analytical Modeling of Cyclic Shear-Flexure Interaction in Reinforced Concrete
Structural Walls”, PhD Dissertation, University of California, Los Angeles.

Joint Elements

1. BeamColumnJoint Element
2. ElasticTubularJoint Element
3. Joint2D Element

BeamColumnJoint Element

This command is used to construct a two-dimensional beam-column-joint element object. The element may be used
with both two-dimensional and three-dimensional structures; however, load is transferred only in the plane of the
element.

52 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

element(’beamColumnJoint’, eleTag, *eleNodes, Mat1Tag, Mat2Tag, Mat3Tag, Mat4Tag, Mat5Tag,

Mat6Tag, Mat7Tag, Mat8Tag, Mat9Tag, Mat10Tag, Mat11Tag, Mat12Tag, Mat13Tag, <eleHeight-
Fac=1.0, eleWidthFac=1.0>)

eleTag unique element object tag

(int)
eleNodes a list of four element nodes
(list (int))
Mat1Tag uniaxial material tag for left bar-slip spring at node 1
(int)
Mat2Tag uniaxial material tag for right bar-slip spring at node 1
(int)
Mat3Tag uniaxial material tag for interface-shear spring at node 1
(int)
Mat4Tag uniaxial material tag for lower bar-slip spring at node 2
(int)
Mat5Tag uniaxial material tag for upper bar-slip spring at node 2
(int)
Mat6Tag uniaxial material tag for interface-shear spring at node 2
(int)
Mat7Tag uniaxial material tag for left bar-slip spring at node 3
(int)
Mat8Tag uniaxial material tag for right bar-slip spring at node 3
(int)
Mat9Tag uniaxial material tag for interface-shear spring at node 3
(int)
Mat10Tag uniaxial material tag for lower bar-slip spring at node 4
(int)
Mat11Tag uniaxial material tag for upper bar-slip spring at node 4
(int)
Mat12Tag uniaxial material tag for interface-shear spring at node 4
(int)
Mat13Tag uniaxial material tag for shear-panel
(int)
eleHeightFac floating point value (as a ratio to the total height of the element) to be considered for de-
(float) termination of the distance in between the tension-compression couples (optional, default:
1.0)
eleWidthFacfloating point value (as a ratio to the total width of the element) to be considered for de-
(float) termination of the distance in between the tension-compression couples (optional, default:
1.0)

See also:
Notes

ElasticTubularJoint Element

This command is used to construct an ElasticTubularJoint element object, which models joint flexibility of tubular
joints in two dimensional analysis of any structure having tubular joints.
element(’ElasticTubularJoint’, eleTag, *eleNodes, Brace_Diameter, Brace_Angle, E, Chord_Diameter,
Chord_Thickness, Chord_Angle)

1.4. Model Commands 53

OpenSeesPy Documentation, Release 3.5.1.3

eleTag (int) unique element object tag

eleNodes (list (int)) a list of two element nodes
Brace_Diameter (float) outer diameter of brace
Brace_Angle (float) angle between brace and chord axis 0 < Brace_Angle < 90
E (float) Young’s Modulus
Chord_Diameter (float) outer diameter of chord
Chord_Thickness (float) thickness of chord
Chord_Angle (float) angle between chord axis and global x-axis 0 < Chord_Angle < 180

See also:
Notes

Joint2D Element

This command is used to construct a two-dimensional beam-column-joint element object. The two dimensional beam-
column joint is idealized as a parallelogram shaped shear panel with adjacent elements connected to its mid-points.
The midpoints of the parallelogram are referred to as external nodes. These nodes are the only analysis components
that connect the joint element to the surrounding structure.
element(’Joint2D’, eleTag, *eleNodes, <Mat1, Mat2, Mat3, Mat4>, MatC, LrgDspTag, <’-damage’, Dmg-
Tag>, <’-damage’, Dmg1 Dmg2 Dmg3 Dmg4 DmgC>)

54 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

eleTag unique element object tag

(int)
eleNodesa list of five element nodes = [nd1,nd2,nd3,nd4,ndC]. ndC is the central node of beam-
(list column joint. (the tag ndC is used to generate the internal node, thus, the node should not exist
(int)) in the domain or be used by any other node)
Mat1 uniaxial material tag for interface rotational spring at node 1. Use a zero tag to indicate the case
(int) that a beam-column element is rigidly framed to the joint. (optional)
Mat2 uniaxial material tag for interface rotational spring at node 2. Use a zero tag to indicate the case
(int) that a beam-column element is rigidly framed to the joint. (optional)
Mat3 uniaxial material tag for interface rotational spring at node 3. Use a zero tag to indicate the case
(int) that a beam-column element is rigidly framed to the joint. (optional)
Mat4 uniaxial material tag for interface rotational spring at node 4. Use a zero tag to indicate the case
(int) that a beam-column element is rigidly framed to the joint. (optional)
MatC uniaxial material tag for rotational spring of the central node that describes shear panel behavior
(int)
an integer indicating the flag for considering large deformations: * 0 - for small deformations
LrgDspTag
(int) and constant geometry * 1 - for large deformations and time varying geometry * 2 - for large
deformations ,time varying geometry and length correction
DmgTag damage model tag
(int)
Dmg1 damage model tag for Mat1
(int)
Dmg2 damage model tag for Mat2
(int)
Dmg3 damage model tag for Mat3
(int)
Dmg4 damage model tag for Mat4
(int)
DmgC panel damage model tag
(int)

See also:
Notes

Link Elements

1. Two Node Link Element

Two Node Link Element

This command is used to construct a twoNodeLink element object, which is defined by two nodes. The element can
have zero or non-zero length. This element can have 1 to 6 degrees of freedom, where only the transverse and rotational
degrees of freedom are coupled as long as the element has non-zero length. In addition, if the element length is larger
than zero, the user can optionally specify how the P-Delta moments around the local x- and y-axis are distributed
among a moment at node i, a moment at node j, and a shear couple. The sum of these three ratios is always equal to
1. In addition the shear center can be specified as a fraction of the element length from the iNode. The element does
not contribute to the Rayleigh damping by default. If the element has non-zero length, the local x-axis is determined
from the nodal geometry unless the optional x-axis vector is specified in which case the nodal geometry is ignored
and the user-defined orientation is utilized. It is important to recognize that if this element has zero length, it does not
consider the geometry as given by the nodal coordinates, but utilizes the user-defined orientation vectors to determine
the directions of the springs.

1.4. Model Commands 55

OpenSeesPy Documentation, Release 3.5.1.3

element(’twoNodeLink’, eleTag, *eleNodes, ’-mat’, *matTags, ’-dir’, *dir, <’-orient’, *vecx, *vecyp>, <’-
pDelta’, *pDeltaVals>, <’-shearDist’, *shearDist>, <’-doRayleigh’>, <’-mass’, m>)

eleTag (int) unique element object tag

eleNodes (list (int)) a list of two element nodes
matTags (list (int)) a list of tags associated with previously-defined Uni-
axialMaterial objects
dirs (list (int)) a list material directions:
• 2D-case: 1 , 2 - translations along local x,y
axes; 3 - rotation about local z axis
• 3D-case: 1, 2, 3 - translations along local x,y,z
axes; 4, 5, 6 - rotations about local x,y,z axes

vecx (list (float)) vector components in global coordinates defining lo-

cal x-axis (optional)
vecyp (list (float)) vector components in global coordinates defining lo-
cal y-axis (optional)
pDeltaVals (list (float)) P-Delta moment contribution ratios, size of ratio
vector is 2 for 2D-case and 4 for 3D-case (en-
tries: [My_iNode, My_jNode, Mz_iNode,
Mz_jNode]) My_iNode + My_jNode <= 1.0,
Mz_iNode + Mz_jNode <= 1.0. Remaining P-
Delta moments are resisted by shear couples. (op-
tional)
sDratios (list (float)) shear distances from iNode as a fraction of the
element length, size of ratio vector is 1 for 2D-
case and 2 for 3D-case. (entries: [dy_iNode,
dz_iNode]) (optional, default = [0.5, 0.5])
'-doRayleigh' (str) to include Rayleigh damping from the element (op-
tional, default = no Rayleigh damping contribution)
m (float) element mass (optional, default = 0.0)

See also:
Notes

Bearing Elements

1. Elastomeric Bearing (Plasticity) Element

2. Elastomeric Bearing (Bouc-Wen) Element
3. Flat Slider Bearing Element
4. Single Friction Pendulum Bearing Element
5. Triple Friction Pendulum Bearing Element
6. Triple Friction Pendulum Element
7. MultipleShearSpring Element
8. KikuchiBearing Element
9. YamamotoBiaxialHDR Element
10. ElastomericX

56 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

11. LeadRubberX
12. HDR
13. RJ-Watson EQS Bearing Element
14. FPBearingPTV

Elastomeric Bearing (Plasticity) Element

This command is used to construct an elastomericBearing element object, which is defined by two nodes. The element
can have zero length or the appropriate bearing height. The bearing has unidirectional (2D) or coupled (3D) plasticity
properties for the shear deformations, and force-deformation behaviors defined by UniaxialMaterials in the remaining
two (2D) or four (3D) directions. By default (sDratio = 0.5) P-Delta moments are equally distributed to the two
end-nodes. To avoid the introduction of artificial viscous damping in the isolation system (sometimes referred to
as “damping leakage in the isolation system”), the bearing element does not contribute to the Rayleigh damping by
default. If the element has non-zero length, the local x-axis is determined from the nodal geometry unless the optional
x-axis vector is specified in which case the nodal geometry is ignored and the user-defined orientation is utilized.
element(’elastomericBearingPlasticity’, eleTag, *eleNodes, kInit, qd, alpha1, alpha2, mu, ’-P’, PMatTag,
’-Mz’, MzMatTag, <’-orient’, x1, x2, x3, y1, y2, y3>, <’-shearDist’, sDratio>, <’-doRayleigh’>,
<’-mass’, m>)
For a two-dimensional problem
element(’elastomericBearingPlasticity’, eleTag, *eleNodes, kInit, qd, alpha1, alpha2, mu, ’-P’, PMatTag,
’-T’, TMatTag, ’-My’, MyMatTag, ’-Mz’, MzMatTag, <’-orient’, <x1, x2, x3>, y1, y2, y3>, <’-
shearDist’, sDratio>, <’-doRayleigh’>, <’-mass’, m>)
For a three-dimensional problem

eleTag (int) unique element object tag

eleNodes (list a list of two element nodes
(int))
kInit (float) initial elastic stiffness in local shear direction
qd (float) characteristic strength
alpha1 (float) post yield stiffness ratio of linear hardening component
alpha2 (float) post yield stiffness ratio of non-linear hardening component
mu (float) exponent of non-linear hardening component
PMatTag (int) tag associated with previously-defined UniaxialMaterial in axial direction
TMatTag (int) tag associated with previously-defined UniaxialMaterial in torsional direction
MyMatTag (int) tag associated with previously-defined UniaxialMaterial in moment direction around
local y-axis
MzMatTag (int) tag associated with previously-defined UniaxialMaterial in moment direction around
local z-axis
x1 x2 x3 (float) vector components in global coordinates defining local x-axis (optional)
y1 y2 y3 (float) vector components in global coordinates defining local y-axis (optional)
sDratio (float) shear distance from iNode as a fraction of the element length (optional, default = 0.5)
'-doRayleigh' to include Rayleigh damping from the bearing (optional, default = no Rayleigh damp-
(str) ing contribution)
m (float) element mass (optional, default = 0.0)

See also:
Notes

1.4. Model Commands 57

OpenSeesPy Documentation, Release 3.5.1.3

Elastomeric Bearing (Bouc-Wen) Element

This command is used to construct an elastomericBearing element object, which is defined by two nodes. The element
can have zero length or the appropriate bearing height. The bearing has unidirectional (2D) or coupled (3D) plasticity
properties for the shear deformations, and force-deformation behaviors defined by UniaxialMaterials in the remaining
two (2D) or four (3D) directions. By default (sDratio = 0.5) P-Delta moments are equally distributed to the two
end-nodes. To avoid the introduction of artificial viscous damping in the isolation system (sometimes referred to
as “damping leakage in the isolation system”), the bearing element does not contribute to the Rayleigh damping by
default. If the element has non-zero length, the local x-axis is determined from the nodal geometry unless the optional
x-axis vector is specified in which case the nodal geometry is ignored and the user-defined orientation is utilized.
element(’ElastomericBearingBoucWen’, eleTag, *eleNodes, kInit, qd, alpha1, alpha2, mu, eta, beta,
gamma, ’-P’, PMatTag, ’-Mz’, MzMatTag, <’-orient’, *orientVals>, <’-shearDist’, shearDist>,
<’-doRayleigh’>, <’-mass’, mass>)
For a two-dimensional problem
element(’ElastomericBearingBoucWen’, eleTag, *eleNodes, kInit, qd, alpha1, alpha2, mu, eta, beta, gamma,
’-P’, PMatTag, ’-T’, TMatTag, ’-My’, MyMatTag, ’-Mz’, MzMatTag, <’-orient’, *orientVals>, <’-
shearDist’, shearDist>, <’-doRayleigh’>, <’-mass’, mass>)
For a three-dimensional problem

eleTag (int) unique element object tag

eleNodes a list of two element nodes
(list (int))
kInit (float) initial elastic stiffness in local shear direction
qd (float) characteristic strength
alpha1 (float) post yield stiffness ratio of linear hardening component
alpha2 (float) post yield stiffness ratio of non-linear hardening component
mu (float) exponent of non-linear hardening component
eta (float) yielding exponent (sharpness of hysteresis loop corners) (default = 1.0)
beta (float) first hysteretic shape parameter (default = 0.5)
gamma (float) second hysteretic shape parameter (default = 0.5)
PMatTag (int) tag associated with previously-defined UniaxialMaterial in axial direction
TMatTag (int) tag associated with previously-defined UniaxialMaterial in torsional direction
MyMatTag tag associated with previously-defined UniaxialMaterial in moment direction around local
(int) y-axis
MzMatTag tag associated with previously-defined UniaxialMaterial in moment direction around local
(int) z-axis
orientVals vector components in global coordinates defining local x-axis (optional), vector compo-
(list (int)) nents in global coordinates defining local y-axis (optional)
shearDist shear distance from iNode as a fraction of the element length (optional, default = 0.5)
(float)
'-doRayleigh'to include Rayleigh damping from the bearing (optional, default = no Rayleigh damping
(str) contribution)
mass (float) element mass (optional, default = 0.0)

See also:
Notes

Flat Slider Bearing Element

This command is used to construct a flatSliderBearing element object, which is defined by two nodes. The iNode rep-
resents the flat sliding surface and the jNode represents the slider. The element can have zero length or the appropriate

58 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

bearing height. The bearing has unidirectional (2D) or coupled (3D) friction properties for the shear deformations,
and force-deformation behaviors defined by UniaxialMaterials in the remaining two (2D) or four (3D) directions. To
capture the uplift behavior of the bearing, the user-specified UniaxialMaterial in the axial direction is modified for
no-tension behavior. By default (sDratio = 0.0) P-Delta moments are entirely transferred to the flat sliding surface (iN-
ode). It is important to note that rotations of the flat sliding surface (rotations at the iNode) affect the shear behavior
of the bearing. To avoid the introduction of artificial viscous damping in the isolation system (sometimes referred to
as “damping leakage in the isolation system”), the bearing element does not contribute to the Rayleigh damping by
default. If the element has non-zero length, the local x-axis is determined from the nodal geometry unless the optional
x-axis vector is specified in which case the nodal geometry is ignored and the user-defined orientation is utilized.
element(’flatSliderBearing’, eleTag, *eleNodes, frnMdlTag, kInit, ’-P’, PMatTag, ’-Mz’, MzMatTag, <’-
orient’, x1, x2, x3, y1, y2, y3>, <’-shearDist’, sDratio>, <’-doRayleigh’>, <’-mass’, m>, <’-
maxIter’, iter, tol>)
For a two-dimensional problem
element(’flatSliderBearing’, eleTag, *eleNodes, frnMdlTag, kInit, ’-P’, PMatTag, ’-T’, TMatTag, ’-My’, My-
MatTag, ’-Mz’, MzMatTag, <’-orient’, <x1, x2, x3>, y1, y2, y3>, <’-shearDist’, sDratio>, <’-
doRayleigh’>, <’-mass’, m>, <’-iter’, maxIter, tol>)
For a three-dimensional problem

eleTag (int) unique element object tag

eleNodes (list a list of two element nodes
(int))
frnMdlTag (float) tag associated with previously-defined FrictionModel
kInit (float) initial elastic stiffness in local shear direction
PMatTag (int) tag associated with previously-defined UniaxialMaterial in axial direction
TMatTag (int) tag associated with previously-defined UniaxialMaterial in torsional direction
MyMatTag (int) tag associated with previously-defined UniaxialMaterial in moment direction around
local y-axis
MzMatTag (int) tag associated with previously-defined UniaxialMaterial in moment direction around
local z-axis
x1 x2 x3 (float) vector components in global coordinates defining local x-axis (optional)
y1 y2 y3 (float) vector components in global coordinates defining local y-axis (optional)
sDratio (float) shear distance from iNode as a fraction of the element length (optional, default = 0.0)
'-doRayleigh' to include Rayleigh damping from the bearing (optional, default = no Rayleigh damp-
(str) ing contribution)
m (float) element mass (optional, default = 0.0)
iter (int) maximum number of iterations to undertake to satisfy element equilibrium (optional,
default = 20)
tol (float) convergence tolerance to satisfy element equilibrium (optional, default = 1E-8)

See also:
Notes

Single Friction Pendulum Bearing Element

This command is used to construct a singleFPBearing element object, which is defined by two nodes. The iNode
represents the concave sliding surface and the jNode represents the articulated slider. The element can have zero
length or the appropriate bearing height. The bearing has unidirectional (2D) or coupled (3D) friction properties (with
post-yield stiffening due to the concave sliding surface) for the shear deformations, and force-deformation behaviors
defined by UniaxialMaterials in the remaining two (2D) or four (3D) directions. To capture the uplift behavior of
the bearing, the user-specified UniaxialMaterial in the axial direction is modified for no-tension behavior. By default
(sDratio = 0.0) P-Delta moments are entirely transferred to the concave sliding surface (iNode). It is important to note

1.4. Model Commands 59

OpenSeesPy Documentation, Release 3.5.1.3

that rotations of the concave sliding surface (rotations at the iNode) affect the shear behavior of the bearing. To avoid
the introduction of artificial viscous damping in the isolation system (sometimes referred to as “damping leakage in
the isolation system”), the bearing element does not contribute to the Rayleigh damping by default. If the element has
non-zero length, the local x-axis is determined from the nodal geometry unless the optional x-axis vector is specified
in which case the nodal geometry is ignored and the user-defined orientation is utilized.
element(’singleFPBearing’, eleTag, *eleNodes, frnMdlTag, Reff, kInit, ’-P’, PMatTag, ’-Mz’, MzMatTag, <’-
orient’, x1, x2, x3, y1, y2, y3>, <’-shearDist’, sDratio>, <’-doRayleigh’>, <’-mass’, m>, <’-iter’,
maxIter, tol>)
For a two-dimensional problem
element(’singleFPBearing’, eleTag, *eleNodes, frnMdlTag, Reff, kInit, ’-P’, PMatTag, ’-T’, TMatTag, ’-My’,
MyMatTag, ’-Mz’, MzMatTag, <’-orient’, <x1, x2, x3>, y1, y2, y3>, <’-shearDist’, sDratio>, <’-
doRayleigh’>, <’-mass’, m>, <’-iter’, maxIter, tol>)
For a three-dimensional problem

eleTag (int) unique element object tag

eleNodes (list a list of two element nodes
(int))
frnMdlTag (float) tag associated with previously-defined FrictionModel
Reff (float) effective radius of concave sliding surface
kInit (float) initial elastic stiffness in local shear direction
PMatTag (int) tag associated with previously-defined UniaxialMaterial in axial direction
TMatTag (int) tag associated with previously-defined UniaxialMaterial in torsional direction
MyMatTag (int) tag associated with previously-defined UniaxialMaterial in moment direction around
local y axis
MzMatTag (int) tag associated with previously-defined UniaxialMaterial in moment direction around
local z-axis
x1 x2 x3 (float) vector components in global coordinates defining local x-axis (optional)
y1 y2 y3 (float) vector components in global coordinates defining local y-axis (optional)
sDratio (float) shear distance from iNode as a fraction of the element length (optional, default = 0.0)
'-doRayleigh' to include Rayleigh damping from the bearing (optional, default = no Rayleigh damp-
(str) ing contribution)
m (float) element mass (optional, default = 0.0)
maxIter (int) maximum number of iterations to undertake to satisfy element equilibrium (optional,
default = 20)
tol (float) convergence tolerance to satisfy element equilibrium (optional, default = 1E-8)

See also:
Notes

Triple Friction Pendulum Bearing Element

This command is used to construct a Triple Friction Pendulum Bearing element object, which is defined by two nodes.
The element can have zero length or the appropriate bearing height. The bearing has unidirectional (2D) or coupled
(3D) friction properties (with post-yield stiffening due to the concave sliding surface) for the shear deformations,
and force-deformation behaviors defined by UniaxialMaterials in the remaining two (2D) or four (3D) directions. To
capture the uplift behavior of the bearing, the user-specified UniaxialMaterial in the axial direction is modified for
no-tension behavior. P-Delta moments are entirely transferred to the concave sliding surface (iNode). It is important
to note that rotations of the concave sliding surface (rotations at the iNode) affect the shear behavior of the bearing.
If the element has non-zero length, the local x-axis is determined from the nodal geometry unless the optional x-axis
vector is specified in which case the nodal geometry is ignored and the user-defined orientation is utilized.

60 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

element(’TFP’, eleTag, *eleNodes, R1, R2, R3, R4, Db1, Db2, Db3, Db4, d1, d2, d3, d4, mu1, mu2, mu3,
mu4, h1, h2, h3, h4, H0, colLoad, <K>)

eleTag (int) unique element object tag

eleNodes (list a list of two element nodes
(int))
R1 (float) Radius of inner bottom sliding surface
R2 (float) Radius of inner top sliding surface
R3 (float) Radius of outer bottom sliding surface
R4 (float) Radius of outer top sliding surface
Db1 (float) Diameter of inner bottom sliding surface
Db2 (float) Diameter of inner top sliding surface
Db3 (float) Diameter of outer bottom sliding surface
Db4 (float) Diameter of outer top sliding surface
d1 (float) diameter of inner slider
d2 (float) diameter of inner slider
d3 (float) diameter of outer bottom slider
d4 (float) diameter of outer top slider
mu1 (float) friction coefficient of inner bottom sliding surface
mu2 (float) friction coefficient of inner top sliding surface
mu3 (float) friction coefficient of outer bottom sliding surface
mu4 (float) friction coefficient of outer top sliding surface
h1 (float) height from inner bottom sliding surface to center of bearing
h2 (float) height from inner top sliding surface to center of bearing
h3 (float) height from outer bottom sliding surface to center of bearing
h4 (float) height from inner top sliding surface to center of bearing
H0 (float) total height of bearing
colLoad (float) initial axial load on bearing (only used for first time step then load come from model)
K (float) optional, stiffness of spring in vertical dirn (dof 2 if ndm= 2, dof 3 if ndm = 3) (de-
fault=1.0e15)

See also:
Notes

Triple Friction Pendulum Element

element(’TripleFrictionPendulum’, eleTag, *eleNodes, frnTag1, frnTag2, frnTag3, vertMatTag, rotZMatTag,

rotXMatTag, rotYMatTag, L1, L2, L3, d1, d2, d3, W, uy, kvt, minFv, tol)

1.4. Model Commands 61

OpenSeesPy Documentation, Release 3.5.1.3

eleTag unique element object tag

(int)
eleNodes a list of two element nodes
(list (int))
frnTag1, = tags associated with previously-defined FrictionModels at the three sliding interfaces
frnTag2
frnTag3
(int)
vertMatTag = Pre-defined material tag for COMPRESSION behavior of the bearing
(int)
rotZMatTag = Pre-defined material tags for rotational behavior about 3-axis, 1-axis and 2-axis, respectively.
rotXMatTag
rotYMatTag
(int)
L1 = effective radii. Li = R_i - h_i (see Figure 1)
L2 L3
(float)
d1 = displacement limits of pendulums (Figure 1). Displacement limit of the bearing is 2 d1 + d2
d2 d3 + d3 + L1. d3/ L3 - L1. d2/ L2
(float)
W (float) = axial force used for the first trial of the first analysis step.
uy = lateral displacement where sliding of the bearing starts. Recommended value = 0.25 to 1 mm.
(float) A smaller value may cause convergence problem.
kvt = Tension stiffness k_vt of the bearing.
(float)
minFv = minimum vertical compression force in the bearing used for computing the horizontal tangent
(>=0) stiffness matrix from the normalized tangent stiffness matrix of the element. minFv is sub-
(float) stituted for the actual compressive force when it is less than minFv, and prevents the element
from using a negative stiffness matrix in the horizontal direction when uplift occurs. The vertical
nodal force returned to nodes is always computed from kvc (or kvt) and vertical deformation,
and thus is not affected by minFv.
tol = relative tolerance for checking the convergence of the element. Recommended value = 1.e-10
(float) to 1.e-3.

See also:
Notes

MultipleShearSpring Element

This command is used to construct a multipleShearSpring (MSS) element object, which is defined by two nodes. This
element consists of a series of identical shear springs arranged radially to represent the isotropic behavior in the local
y-z plane.
element(’multipleShearSpring’, eleTag, *eleNodes, nSpring, ’-mat’, matTag, <’-lim’, lim>, <’-orient’, <x1,
x2, x3>, yp1, yp2, yp3>, <’-mass’, mass>)

62 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

eleTag (int) unique element object tag

eleNodes (list a list of two element nodes
(int))
nSpring (int) number of springs
matTag (int) tag associated with previously-defined UniaxialMaterial object
lim (float) minimum deformation to calculate equivalent coefficient (see note 1)
x1 x2 x3 (float) vector components in global coordinates defining local x-axis
yp1 yp2 yp3 vector components in global coordinates defining vector yp which lies in the local x-y
(float) plane for the element
mass (float) element mass

Note: If dsp is positive and the shear deformation of MSS exceeds dsp, this element calculates equivalent coefficient
to adjust force and stiffness of MSS. The adjusted MSS force and stiffness reproduce the behavior of the previously
defined uniaxial material under monotonic loading in every direction. If dsp is zero, the element does not calculate
the equivalent coefficient.

See also:
Notes

KikuchiBearing Element

This command is used to construct a KikuchiBearing element object, which is defined by two nodes. This element
consists of multiple shear spring model (MSS) and multiple normal spring model (MNS).
element(’KikuchiBearing’, eleTag, *eleNodes, ’-shape’, shape, ’-size’, size, totalRubber, <’-totalHeight’,
totalHeight>, ’-nMSS’, nMSS, ’-matMSS’, matMSSTag, <’-limDisp’, limDisp>, ’-nMNS’, nMNS,
’-matMNS’, matMNSTag, <’-lambda’, lambda>, <’-orient’, <x1, x2, x3>, yp1, yp2, yp3>, <’-
mass’, m>, <’-noPDInput’>, <’-noTilt’>, <’-adjustPDOutput’, ci, cj>, <’-doBalance’, limFo,
limFi, nIter>)

1.4. Model Commands 63

OpenSeesPy Documentation, Release 3.5.1.3

eleTag (int) unique element object tag

eleNodes a list of two element nodes
(list (int))
shape (float) following shapes are available: round, square
size (float) diameter (round shape), length of edge (square shape)
totalRubber total rubber thickness
(float)
totalHeight total height of the bearing (defaulut: distance between iNode and jNode)
(float)
nMSS (int) number of springs in MSS = nMSS
matMSSTag matTag for MSS
(int)
limDisp minimum deformation to calculate equivalent coefficient of MSS (see note 1)
(float)
nMNS (int) number of springs in MNS = nMNS*nMNS (for round and square shape)
matMNSTag matTag for MNS
(int)
lambda (float) parameter to calculate compression modulus distribution on MNS (see note 2)
x1 x2 x3 vector components in global coordinates defining local x-axis
(float)
yp1 yp2 yp3 vector components in global coordinates defining vector yp which lies in the local x-y
(float) plane for the element
m (float) element mass
'-noPDInput' not consider P-Delta moment
(str)
'-noTilt' not consider tilt of rigid link
(str)
ci cj (float) P-Delta moment adjustment for reaction force (default: ci =0.5, cj =0.5)
limFo limFi tolerance of external unbalanced force ( limFo), tolorance of internal unbalanced force
nIter (float) ( limFi), number of iterations to get rid of internal unbalanced force ( nIter)

See also:
Notes

YamamotoBiaxialHDR Element

This command is used to construct a YamamotoBiaxialHDR element object, which is defined by two nodes. This
element can be used to represent the isotropic behavior of high-damping rubber bearing in the local y-z plane.
element(’YamamotoBiaxialHDR’, eleTag, *eleNodes, Tp, DDo, DDi, Hr, <’-coRS‘, cr, cs>, <’-orient‘,
*vecx, *vecyp>, <’-mass‘, m>)

64 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

eleTag (int) unique element object tag

eleNodes (list a list of two element nodes
(int))
Tp (int) compound type = 1 : X0.6R manufactured by Bridgestone corporation.
DDo (float) outer diameter [m]
DDi (float) bore diameter [m]
Hr (float) total thickness of rubber layer [m] Optional Data
cr cs (float) coefficients for shear stress components of tau_r and tau_s
vecx (list a list of vector components in global coordinates defining local x-axis (optional)
(float))
vecyp (list a list of vector components in global coordinates defining vector yp which lies in the
(float)) local x-y plane for the element.
m (float) element mass [kg]

See also:
Notes

ElastomericX

This command is used to construct an ElastomericX bearing element object in three-dimension. The 3D continuum
geometry of an elastomeric bearing is modeled as a 2-node, 12 DOF discrete element. This elements extends the
formulation of Elastomeric_Bearing_(Bouc-Wen)_Element element. However, instead of the user providing material
models as input arguments, it only requires geometric and material properties of an elastomeric bearing as arguments.
The material models in six direction are formulated within the element from input arguments. The time-dependent val-
ues of mechanical properties (e.g., shear stiffness, buckling load capacity) can also be recorded using the “parameters”
recorder.
element(’ElastomericX’, eleTag, *eleNodes, Fy, alpha, Gr, Kbulk, D1, D2, ts, tr, n, <<x1, x2, x3>, y1, y2,
y3>, <kc>, <PhiM>, <ac>, <sDratio>, <m>, <cd>, <tc>, <tag1>, <tag2>, <tag3>, <tag4>)
For 3D problem

1.4. Model Commands 65

OpenSeesPy Documentation, Release 3.5.1.3

eleTag (int) unique element object tag

eleNodes (list a list of two element nodes
(int))
Fy (float) yield strength
alpha (float) post-yield stiffness ratio
Gr (float) shear modulus of elastomeric bearing
Kbulk (float) bulk modulus of rubber
D1 (float) internal diameter
D2 (float) outer diameter (excluding cover thickness)
ts (float) single steel shim layer thickness
tr (float) single rubber layer thickness
n (int) number of rubber layers
x1 x2 x3 (float) vector components in global coordinates defining local x-axis (optional)
y1 y2 y3 (float) vector components in global coordinates defining local y-axis (optional)
kc (float) cavitation parameter (optional, default = 10.0)
PhiM (float) damage parameter (optional, default = 0.5)
ac (float) strength reduction parameter (optional, default = 1.0)
sDratio (float) shear distance from iNode as a fraction of the element length (optional, default =
0.5)
m (float) element mass (optional, default = 0.0)
cd (float) viscous damping parameter (optional, default = 0.0)
tc (float) cover thickness (optional, default = 0.0)
tag1 (float) Tag to include cavitation and post-cavitation (optional, default = 0)
tag2 (float) Tag to include buckling load variation (optional, default = 0)
tag3 (float) Tag to include horizontal stiffness variation (optional, default = 0)
tag4 (float) Tag to include vertical stiffness variation (optional, default = 0)

Note: Because default values of heating parameters are in SI units, user must override the default heating parameters
values if using Imperial units
User should distinguish between yield strength of elastomeric bearing (𝐹𝑦 ) and characteristic strength (𝑄𝑑 ): 𝑄𝑑 =
𝐹𝑦 * (1 − 𝑎𝑙𝑝ℎ𝑎)

See also:
Notes

LeadRubberX

This command is used to construct a LeadRubberX bearing element object in three-dimension. The 3D continuum
geometry of a lead rubber bearing is modeled as a 2-node, 12 DOF discrete element. It extends the formulation of
ElastomericX by including strength degradation in lead rubber bearing due to heating of the lead-core. The Lead-
RubberX element requires only the geometric and material properties of an elastomeric bearing as arguments. The
material models in six direction are formulated within the element from input arguments. The time-dependent values
of mechanical properties (e.g., shear stiffness, buckling load capacity, temperature in the lead-core, yield strength) can
also be recorded using the “parameters” recorder.
element(’LeadRubberX’, eleTag, *eleNodes, Fy, alpha, Gr, Kbulk, D1, D2, ts, tr, n, <<x1, x2, x3>, y1, y2,
y3>, <kc>, <PhiM>, <ac>, <sDratio>, <m>, <cd>, <tc>, <qL>, <cL>, <kS>, <aS>, <tag1>,
<tag2>, <tag3>, <tag4>, <tag5>)

66 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

eleTag (int) unique element object tag

eleNodes (list a list of two element nodes
(int))
Fy (float) yield strength
alpha (float) post-yield stiffness ratio
Gr (float) shear modulus of elastomeric bearing
Kbulk (float) bulk modulus of rubber
D1 (float) internal diameter
D2 (float) outer diameter (excluding cover thickness)
ts (float) single steel shim layer thickness
tr (float) single rubber layer thickness
n (int) number of rubber layers
x1 x2 x3 (float) vector components in global coordinates defining local x-axis (optional)
y1 y2 y3 (float) vector components in global coordinates defining local y-axis (optional)
kc (float) cavitation parameter (optional, default = 10.0)
PhiM (float) damage parameter (optional, default = 0.5)
ac (float) strength reduction parameter (optional, default = 1.0)
sDratio (float) shear distance from iNode as a fraction of the element length (optional, default = 0.5)
m (float) element mass (optional, default = 0.0)
cd (float) viscous damping parameter (optional, default = 0.0)
tc (float) cover thickness (optional, default = 0.0)
qL (float) density of lead (optional, default = 11200 kg/m3)
cL (float) specific heat of lead (optional, default = 130 N-m/kg oC)
kS (float) thermal conductivity of steel (optional, default = 50 W/m oC)
aS (float) thermal diffusivity of steel (optional, default = 1.41e-05 m2/s)
tag1 (int) Tag to include cavitation and post-cavitation (optional, default = 0)
tag2 (int) Tag to include buckling load variation (optional, default = 0)
tag3 (int) Tag to include horizontal stiffness variation (optional, default = 0)
tag4 (int) Tag to include vertical stiffness variation (optional, default = 0)
tag5 (int) Tag to include strength degradation in shear due to heating of lead core (optional,
default = 0)

Note: Because default values of heating parameters are in SI units, user must override the default heating parameters
values if using Imperial units
User should distinguish between yield strength of elastomeric bearing (𝐹𝑦 ) and characteristic strength (𝑄𝑑 ): 𝑄𝑑 =
𝐹𝑦 * (1 − 𝑎𝑙𝑝ℎ𝑎)

See also:
Notes

HDR

This command is used to construct an HDR bearing element object in three-dimension. The 3D continuum geometry
of an high damping rubber bearing is modeled as a 2-node, 12 DOF discrete element. This is the third element in the
series of elements developed for analysis of base-isolated structures under extreme loading (others being ElastomericX
and LeadRubberX). The major difference between HDR element with ElastomericX is the hysteresis model in shear.
The HDR element uses a model proposed by Grant et al. (2004) to capture the shear behavior of a high damping
rubber bearing. The time-dependent values of mechanical properties (e.g., vertical stiffness, buckling load capacity)
can also be recorded using the “parameters” recorder.

1.4. Model Commands 67

OpenSeesPy Documentation, Release 3.5.1.3

element(’HDR’, eleTag, *eleNodes, Gr, Kbulk, D1, D2, ts, tr, n, a1, a2, a3, b1, b2, b3, c1, c2, c3, c4, <<x1,
x2, x3>, y1, y2, y3>, <kc>, <PhiM>, <ac>, <sDratio>, <m>, <tc>)
For 3D problem

eleTag (int) unique element object tag

eleNodes (list (int)) a list of two element nodes
Gr (float) shear modulus of elastomeric bearing
Kbulk (float) bulk modulus of rubber
D1 (float) internal diameter
D2 (float) outer diameter (excluding cover thickness)
ts (float) single steel shim layer thickness
tr (float) single rubber layer thickness
n (int) number of rubber layers
a1 a2 a3 b1 b2 b3 c1 c2 c3 c4 parameters of the Grant model
(float)
x1 x2 x3 (float) vector components in global coordinates defining local x-axis (op-
tional)
y1 y2 y3 (float) vector components in global coordinates defining local y-axis (op-
tional)
kc (float) cavitation parameter (optional, default = 10.0)
PhiM (float) damage parameter (optional, default = 0.5)
ac (float) strength reduction parameter (optional, default = 1.0)
sDratio (float) shear distance from iNode as a fraction of the element length (optional,
default = 0.5)
m (float) element mass (optional, default = 0.0)
tc (float) cover thickness (optional, default = 0.0)

See also:
Notes

RJ-Watson EQS Bearing Element

This command is used to construct a RJWatsonEqsBearing element object, which is defined by two nodes. The iNode
represents the masonry plate and the jNode represents the sliding surface plate. The element can have zero length or
the appropriate bearing height. The bearing has unidirectional (2D) or coupled (3D) friction properties (with post-
yield stiffening due to the mass-energy-regulator (MER) springs) for the shear deformations, and force-deformation
behaviors defined by UniaxialMaterials in the remaining two (2D) or four (3D) directions. To capture the uplift
behavior of the bearing, the user-specified UniaxialMaterial in the axial direction is modified for no-tension behavior.
By default (sDratio = 1.0) P-Delta moments are entirely transferred to the sliding surface (jNode). It is important to
note that rotations of the sliding surface (rotations at the jNode) affect the shear behavior of the bearing. To avoid the
introduction of artificial viscous damping in the isolation system (sometimes referred to as “damping leakage in the
isolation system”), the bearing element does not contribute to the Rayleigh damping by default. If the element has
non-zero length, the local x-axis is determined from the nodal geometry unless the optional x-axis vector is specified
in which case the nodal geometry is ignored and the user-defined orientation is utilized.
element(’RJWatsonEqsBearing’, eleTag, *eleNodes, frnMdlTag, kInit, ’-P’, PMatTag, ’-Vy’, VyMatTag, ’-
Mz’, MzMatTag, <’-orient’, x1, x2, x3, y1, y2, y3>, <’-shearDist’, sDratio>, <’-doRayleigh’>,
<’-mass’, m>, <’-iter’, maxIter, tol>)
For a two-dimensional problem
element(’RJWatsonEqsBearing’, eleTag, *eleNodes, frnMdlTag, kInit, ’-P’, PMatTag, ’-Vy’, VyMatTag, ’-
Vz’, VzMatTag, ’-T’, TMatTag, ’-My’, MyMatTag, ’-Mz’, MzMatTag, <’-orient’, <x1, x2, x3>, y1,
y2, y3>, <’-shearDist’, sDratio>, <’-doRayleigh’>, <’-mass’, m>, <’-iter’, maxIter, tol>)

68 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

For a three-dimensional problem

eleTag (int) unique element object tag

eleNodes a list of two element nodes
(list (int))
frnMdlTag tag associated with previously-defined FrictionModel
(float)
kInit (float) initial stiffness of sliding friction component in local shear direction
'-P' tag associated with previously-defined UniaxialMaterial in axial direction
PMatTag
(int)
'-Vy' tag associated with previously-defined UniaxialMaterial in shear direction along local
VyMatTag y-axis (MER spring behavior not including friction)
(int)
'-Vz' tag associated with previously-defined UniaxialMaterial in shear direction along local
VzMatTag z-axis (MER spring behavior not including friction)
(int)
'-T' tag associated with previously-defined UniaxialMaterial in torsional direction
TMatTag
(int)
'-My' tag associated with previously-defined UniaxialMaterial in moment direction around local
MyMatTag y-axis
(int)
'-Mz' tag associated with previously-defined UniaxialMaterial in moment direction around local
MzMatTag z-axis
(int)
x1 x2 x3 vector components in global coordinates defining local x-axis (optional)
(float)
y1 y2 y3 vector components in global coordinates defining local y-axis (optional)
(float)
sDratio shear distance from iNode as a fraction of the element length (optional, default = 0.0)
(float)
'-doRayleigh'to include Rayleigh damping from the bearing (optional, default = no Rayleigh damping
(str) contribution)
m (float) element mass (optional, default = 0.0)
maxIter (int) maximum number of iterations to undertake to satisfy element equilibrium (optional, de-
fault = 20)
tol (float) convergence tolerance to satisfy element equilibrium (optional, default = 1E-8)

See also:
Notes

FPBearingPTV

The FPBearingPTV command creates a single Friction Pendulum bearing element, which is capable of accounting for
the changes in the coefficient of friction at the sliding surface with instantaneous values of the sliding velocity, axial
pressure and temperature at the sliding surface. The constitutive modelling is similar to the existing singleFPBear-
ing element, otherwise. The FPBearingPTV element has been verified and validated in accordance with the ASME
guidelines, details of which are presented in Chapter 4 of Kumar et al. (2015a).

1.4. Model Commands 69

OpenSeesPy Documentation, Release 3.5.1.3

element(’FPBearingPTV’, eleTag, *eleNodes, MuRef, IsPressureDependent, pRef, IsTemperatureDepen-

dent, Diffusivity, Conductivity, IsVelocityDependent, rateParameter, ReffectiveFP, Radius_Contact,
kInitial, theMaterialA, theMaterialB, theMaterialC, theMaterialD, x1, x2, x3, y1, y2, y3, shearDist,
doRayleigh, mass, iter, tol, unit)

eleTag (int) unique element object tag

eleNodes (list (int)) a list of two element nodes
MuRef (float) Reference coefficient of friction
IsPressureDependent (int) 1 if the coefficient of friction is a function of instan-
taneous axial pressure
pRef (float) Reference axial pressure (the bearing pressure under
static loads)
IsTemperatureDependent (int) 1 if the coefficient of friction is a function of instan-
taneous temperature at the sliding surface
Diffusivity (float) Thermal diffusivity of steel
Conductivity (float) Thermal conductivity of steel
IsVelocityDependent (int) 1 if the coefficient of friction is a function of instan-
taneous velocity at the sliding surface
rateParameter (float) The exponent that determines the shape of the coef-
ficient of friction vs. sliding velocity curve
ReffectiveFP (float) Effective radius of curvature of the sliding surface of
the FPbearing
Radius_Contact (float) Radius of contact area at the sliding surface
kInitial (float) Lateral stiffness of the sliding bearing before sliding
begins
theMaterialA (int) Tag for the uniaxial material in the axial direction
theMaterialB (int) Tag for the uniaxial material in the torsional direction
theMaterialC (int) Tag for the uniaxial material for rocking about local
Y axis
theMaterialD (int) Tag for the uniaxial material for rocking about local
Z axis
x1 x2 x3 (float) Vector components to define local X axis
y1 y2 y3 (float) Vector components to define local Y axis
shearDist (float) Shear distance from iNode as a fraction of the length
of the element
doRayleigh (int) To include Rayleigh damping from the bearing
mass (float) Element mass
iter (int) Maximum number of iterations to satisfy the equilib-
rium of element
tol (float) Convergence tolerance to satisfy the equilibrium of
the element
unit (int) Tag to identify the unit from the list below.
• 1: N, m, s, C
• 2: kN, m, s, C
• 3: N, mm, s, C
• 4: kN, mm, s, C
• 5: lb, in, s, C
• 6: kip, in, s, C
• 7: lb, ft, s, C
• 8: kip, ft, s, C

70 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

See also:
Notes

Quadrilateral Elements

1. Quad Element
2. Shell Element
3. ShellDKGQ
4. ShellDKGT
5. ShellNLDKGQ
6. ShellNLDKGT
7. ShellNL
8. Bbar Plane Strain Quadrilateral Element
9. Enhanced Strain Quadrilateral Element
10. SSPquad Element
11. MVLEM_3D - 3-D MVLEM Element for Flexure-Dominated RC Walls
12. SFI_MVLEM_3D - 3-D Shear-Flexure-Interaction Element for RC Walls

Quad Element

This command is used to construct a FourNodeQuad element object which uses a bilinear isoparametric formulation.
element(’quad’, eleTag, *eleNodes, thick, type, matTag, <pressure=0.0, rho=0.0, b1=0.0, b2=0.0>)

eleTag (int) unique element object tag

eleNodes (list a list of four element nodes in counter-clockwise order
(int))
thick (float) element thickness
type (str) string representing material behavior. The type parameter can be either
'PlaneStrain' or 'PlaneStress'
matTag (int) tag of nDMaterial
pressure surface pressure (optional, default = 0.0)
(float)
rho (float) element mass density (per unit volume) from which a lumped element mass matrix is
computed (optional, default=0.0)
b1 b2 (float) constant body forces defined in the isoparametric domain (optional, default=0.0)

Note:
1. Consistent nodal loads are computed from the pressure and body forces.
2. The valid queries to a Quad element when creating an ElementRecorder object are ‘forces’, ‘stresses,’ and
‘material $matNum matArg1 matArg2 . . . ’ Where $matNum refers to the material object at the integration
point corresponding to the node numbers in the isoparametric domain.

1.4. Model Commands 71

OpenSeesPy Documentation, Release 3.5.1.3

See also:
Notes

Shell Element

This command is used to construct a ShellMITC4 element object, which uses a bilinear isoparametric formulation in
combination with a modified shear interpolation to improve thin-plate bending performance.
element(’ShellMITC4’, eleTag, *eleNodes, secTag)

eleTag unique element object tag

(int)
eleNodes a list of four element nodes in counter-clockwise order
(list (int))
secTag tag associated with previously-defined SectionForceDeformation object. Currently must be
(int) either a 'PlateFiberSection', or 'ElasticMembranePlateSection'

Note:
1. The valid queries to a Quad element when creating an ElementRecorder object are ‘forces’, ‘stresses,’ and
‘material $matNum matArg1 matArg2 . . . ’ Where $matNum refers to the material object at the integration
point corresponding to the node numbers in the isoparametric domain.
2. It is a 3D element with 6 dofs and CAN NOT be used in 2D domain.

See also:
Notes

ShellDKGQ

This command is used to construct a ShellDKGQ element object, which is a quadrilateral shell element based on the
theory of generalized conforming element.
element(’ShellDKGQ’, eleTag, *eleNodes, secTag)

eleTag unique element object tag

(int)
eleNodes a list of four element nodes in counter-clockwise order
(list (int))
secTag tag associated with previously-defined SectionForceDeformation object. Currently can
(int) be a 'PlateFiberSection', a 'ElasticMembranePlateSection' and a
'LayeredShell' section

See also:
Notes

72 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

ShellDKGT

This command is used to construct a ShellDKGT element object, which is a triangular shell element based on the
theory of generalized conforming element.
element(’ShellDKGT’, eleTag, *eleNodes, secTag)

eleTag unique element object tag

(int)
eleNodes a list of three element nodes in clockwise or counter-clockwise order
(list (int))
secTag tag associated with previously-defined SectionForceDeformation object. currently can
(int) be a 'PlateFiberSection', a 'ElasticMembranePlateSection' and a
'LayeredShell' section

See also:
Notes

ShellNLDKGQ

This command is used to construct a ShellNLDKGQ element object accounting for the geometric nonlinearity of large
deformation using the updated Lagrangian formula, which is developed based on the ShellDKGQ element.
element(’ShellNLDKGQ’, eleTag, *eleNodes, secTag)

eleTag unique element object tag

(int)
eleNodes a list of four element nodes in counter-clockwise order
(list (int))
secTag tag associated with previously-defined SectionForceDeformation object. currently can
(int) be a 'PlateFiberSection', a 'ElasticMembranePlateSection' and a
'LayeredShell' section

See also:
Notes

ShellNLDKGT

This command is used to construct a ShellNLDKGT element object accounting for the geometric nonlinearity of large
deformation using the updated Lagrangian formula, which is developed based on the ShellDKGT element.
element(’ShellNLDKGT’, eleTag, *eleNodes, secTag)

1.4. Model Commands 73

OpenSeesPy Documentation, Release 3.5.1.3

eleTag unique element object tag

(int)
eleNodes a list of three element nodes in clockwise or counter-clockwise order around the element
(list (int))
secTag tag associated with previously-defined SectionForceDeformation object. currently can
(int) be a 'PlateFiberSection', a 'ElasticMembranePlateSection' and a
'LayeredShell' section

See also:
Notes

ShellNL

element(’ShellNL’, eleTag, *eleNodes, secTag)

eleTag unique element object tag

(int)
eleNodes a list of nine element nodes, input is the typical, firstly four corner nodes counter-clockwise,
(list (int)) then mid-side nodes counter-clockwise and finally the central node
secTag tag associated with previously-defined SectionForceDeformation object. currently can
(int) be a 'PlateFiberSection', a 'ElasticMembranePlateSection' and a
'LayeredShell' section

See also:
Notes

Bbar Plane Strain Quadrilateral Element

This command is used to construct a four-node quadrilateral element object, which uses a bilinear isoparametric
formulation along with a mixed volume/pressure B-bar assumption. This element is for plane strain problems only.
element(’bbarQuad’, eleTag, *eleNodes, thick, matTag)

eleTag (int) unique element object tag

eleNodes (list (int)) a list of four element nodes in counter-clockwise order
thick (float) element thickness
matTag (int) tag of nDMaterial

Note:
1. PlainStrain only.
2. The valid queries to a Quad element when creating an ElementRecorder object are ‘forces’, ‘stresses,’ and
‘material $matNum matArg1 matArg2 . . . ’ Where $matNum refers to the material object at the integration
point corresponding to the node numbers in the isoparametric domain.

74 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

See also:
Notes

Enhanced Strain Quadrilateral Element

This command is used to construct a four-node quadrilateral element, which uses a bilinear isoparametric formulation
with enhanced strain modes.
element(’enhancedQuad’, eleTag, *eleNodes, thick, type, matTag)

eleTag unique element object tag

(int)
eleNodes a list of four element nodes in counter-clockwise order
(list (int))
thick element thickness
(float)
type string representing material behavior. Valid options depend on the NDMaterial object and
(str) its available material formulations. The type parameter can be either 'PlaneStrain' or
'PlaneStress'
matTag tag of nDMaterial
(int)

See also:
Notes

SSPquad Element

This command is used to construct a SSPquad element object.

element(’SSPquad’, eleTag, *eleNodes, matTag, type, thick, <b1, b2>)

eleTag (int) unique element object tag

eleNodes (list a list of four element nodes in counter-clockwise order
(int))
thick (float) thickness of the element in out-of-plane direction
type (str) string to relay material behavior to the element, can be either 'PlaneStrain' or
'PlaneStress'
matTag (int) unique integer tag associated with previously-defined nDMaterial object
b1 b2 (float) constant body forces in global x- and y-directions, respectively (optional, default = 0.0)

The SSPquad element is a four-node quadrilateral element using physically stabilized single-point integration
(SSP –> Stabilized Single Point). The stabilization incorporates an assumed strain field in which the volu-
metric dilation and the shear strain associated with the the hourglass modes are zero, resulting in an element
which is free from volumetric and shear locking. The elimination of shear locking results in greater coarse
mesh accuracy in bending dominated problems, and the elimination of volumetric locking improves accuracy

1.4. Model Commands 75

OpenSeesPy Documentation, Release 3.5.1.3

in nearly-incompressible problems. Analysis times are generally faster than corresponding full integration el-
ements. The formulation for this element is identical to the solid phase portion of the SSPquadUP element as
described by McGann et al. (2012).

Note:
1. Valid queries to the SSPquad element when creating an ElementalRecorder object correspond to those for the
nDMaterial object assigned to the element (e.g., ‘stress’, ‘strain’). Material response is recorded at the single
integration point located in the center of the element.
2. The SSPquad element was designed with intentions of duplicating the functionality of the Quad Element. If
an example is found where the SSPquad element cannot do something that works for the Quad Element, e.g.,
material updating, please contact the developers listed below so the bug can be fixed.

See also:
Notes

MVLEM_3D - 3-D MVLEM Element for Flexure-Dominated RC Walls

Developed and implemented by:

Kristijan Kolozvari (CSU Fullerton)
Kamiar Kalbasi (CSU Fullerton)
Kutay Orakcal (Bogazici University)
John Wallace (UCLA)

The MVLEM_3D model (Figure 1a) is a three-dimensional four-node element with 24 DOFs for nonlinear analy-
sis of flexure-controlled non-rectangular reinforced concrete walls subjected to multidirectional loading. The model
is an extension of the two-dimensional, two-node Multiple-Vertical-Line-Element-Model (MVLEM). The baseline
MVLEM, which is essentially a line element for rectangular walls subjected to in-plane loading, is extended to a
three-dimensional model formulation by: 1) applying geometric transformation of the element in-plane degrees of
freedom that convert it into a four-node element formulation (Figure 1b), as well as by incorporating linear elastic
out-of-plane behavior based on the Kirchhoff plate theory (Figure 1c). The in-plane and the out-of-plane element
behaviors are uncoupled in the present model.
This element shall be used in Domain defined with -ndm 3 -ndf 6.
element(’MVLEM_3D’, eleTag, *eleNodes, m, ’-thick’, *thick, ’-width’, *widths, ’-rho’, *rho, ’-
matConcrete’, *matConcreteTags, ’-matSteel’, *matSteelTags, ’-matShear’, matShearTag, <’-
CoR’, c>, <’-ThickMod’, tMod>, <’-Poisson’, Nu>, <’-Density’, Dens>)

76 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

Fig. 1: Figure 1: MVLEM_3D Element Formulation

eleTag (int) unique element object tag

eleNodes (list (int)) a list of four element nodes defined in the counter-clockwise direction
m (int) number of element uniaxial fibers
thick (list (float)) a list of m macro-fiber thicknesses
widths (list (float)) a list of m macro-fiber widths
rho (list (float)) a list of m reinforcing ratios corresponding to macro-fibers; for each fiber: 𝑟ℎ𝑜𝑖 =
𝐴𝑠,𝑖 /𝐴𝑔𝑟𝑜𝑠𝑠,𝑖 (1 < 𝑖 < 𝑚)
matConcreteTags a list of m uniaxialMaterial tags for concrete
(list (int))
matSteelTags (list a list of m uniaxialMaterial tags for steel
(int))
matShearTag (int) tag of uniaxialMaterial for shear material
c (float) location of center of rotation from the base (optional; default = 0.4 (recom-
mended))
tMod (float) thickness multiplier (optional; default = 0.63 equivalent to 0.25Ig for out-of-plane
bending)
Nu (float) Poisson ratio for out-of-plane bending (optional; default = 0.25)
Dens (float) density (optional; default = 0.0)

See also:
More information available HERE and in the following reference:
K. Kolozvari, K. Kalbasi, K. Orakcal & J. W. Wallace, “Three-Dimensional Model for Nonlinear Analysis of
Slender Flanged Reinforced Concrete Walls”, Engineering Structures, Volume 236, 1 June 2021, 112105.

SFI_MVLEM_3D - 3-D Shear-Flexure-Interaction Element for RC Walls

Developed and implemented by:

Kristijan Kolozvari (CSU Fullerton)
Kamiar Kalbasi (CSU Fullerton)
Kutay Orakcal (Bogazici University)
John Wallace (UCLA)

1.4. Model Commands 77

OpenSeesPy Documentation, Release 3.5.1.3

The SFI-MVLEM-3D model (Figure 1a) is a three-dimensional four-node element with 24 DOFs that incorporates
axial-flexural-shear interaction and can be used for nonlinear analysis of non-rectangular reinforced concrete walls
subjected to multidirectional loading. The SFI-MVLEM-3D model is an extension of the two-dimensional, two-
node Shear-Flexure-Interaction Multiple-Vertical-Line-Element-Model (SFI-MVLEM). The baseline SFI-MVLEM,
which is essentially a line element for rectangular walls subjected to in-plane loading, is extended in this study to a
three-dimensional model formulation by applying geometric transformation of the element degrees of freedom that
converted it into a four-node element formulation (Figure 1b), as well as by incorporating linear elastic out-of-plane
behavior based on the Kirchhoff plate theory (Figure 1c). The in-plane and the out-of-plane element behaviors are
uncoupled in the present model.
This element shall be used in Domain defined with -ndm 3 -ndf 6.

Fig. 2: Figure 1: SFI_MVLEM_3D Element Formulation

element(’SFI_MVLEM_3D’, eleTag, *eleNodes, m, ’-thick’, *thicks, ’-width’, *widths, ’-mat’, *matTags,

<’-CoR’, c>, <’-ThickMod’, tMod>, <’-Poisson’, Nu>, <’-Density’, Dens>)

eleTag (int) unique element object tag

eleNodes (list a list of four element nodes defined in the counter-clockwise direction
(int))
m (int) number of element uniaxial fibers
thicks (list a list of m macro-fiber thicknesses
(float))
widths (list a list of m macro-fiber widths
(float))
matTags (list (int)) a list of m material tags corresponding to nDmaterial FSAM
c (float) location of center of rotation from the base (optional; default = 0.4 (recommended))
tMod (float) thickness multiplier (optional; default = 0.63 equivalent to 0.25Ig for out-of-plane
bending)
Nu (float) Poisson ratio for out-of-plane bending (optional; default = 0.25)
Dens (float) density (optional; default = 0.0)

See also:
More information available HERE and in the following reference:

78 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

K. Kolozvari, K. Kalbasi, K. Orakcal & J. W. Wallace, “Three-Dimensional Shear-Flexure Interaction Model for
Analysis of Non-Planar Reinforced Concrete Walls”, Journal of Building Engineering, Volume 44, December
2021, 102946.

Triangular Elements

1. Tri31 Element

Tri31 Element

This command is used to construct a constant strain triangular element (Tri31) which uses three nodes and one inte-
gration points.
element(’Tri31’, eleTag, *eleNodes, thick, type, matTag, <pressure, rho, b1, b2>)

eleTag (int) unique element object tag

eleNodes (list a list of three element nodes in counter-clockwise order
(int))
thick (float) element thickness
type (str) string representing material behavior. The type parameter can be either
'PlaneStrain' or 'PlaneStress'
matTag (int) tag of nDMaterial
pressure surface pressure (optional, default = 0.0)
(float)
rho (float) element mass density (per unit volume) from which a lumped element mass matrix is
computed (optional, default=0.0)
b1 b2 (float) constant body forces defined in the domain (optional, default=0.0)

Note:
1. Consistent nodal loads are computed from the pressure and body forces.
2. The valid queries to a Tri31 element when creating an ElementRecorder object are ‘forces’, ‘stresses,’ and
‘material $matNum matArg1 matArg2 . . . ’ Where $matNum refers to the material object at the integration
point corresponding to the node numbers in the domain.

See also:
Notes

Brick Elements

1. Standard Brick Element

2. Bbar Brick Element
3. Twenty Node Brick Element
4. SSPbrick Element

1.4. Model Commands 79

OpenSeesPy Documentation, Release 3.5.1.3

Standard Brick Element

This element is used to construct an eight-node brick element object, which uses a trilinear isoparametric formulation.
element(’stdBrick’, eleTag, *eleNodes, matTag, <b1, b2, b3>)

eleTag (int) unique element object tag

eleNodes (list (int)) a list of eight element nodes in bottom and top faces and in counter-clockwise order
matTag (int) tag of nDMaterial
b1 b2 b3 (float) body forces in global x,y,z directions

Note:
1. The valid queries to a Brick element when creating an ElementRecorder object are ‘forces’, ‘stresses,’ (‘strains’
version > 2.2.0) and ‘material $matNum matArg1 matArg2 . . . ’ Where $matNum refers to the material object
at the integration point corresponding to the node numbers in the isoparametric domain.
2. This element can only be defined in -ndm 3 -ndf 3

See also:
Notes

Bbar Brick Element

This command is used to construct an eight-node mixed volume/pressure brick element object, which uses a trilinear
isoparametric formulation.
element(’bbarBrick’, eleTag, *eleNodes, matTag, <b1, b2, b3>)

eleTag (int) unique element object tag

eleNodes (list (int)) a list of eight element nodes in bottom and top faces and in counter-clockwise order
matTag (int) tag of nDMaterial
b1 b2 b3 (float) body forces in global x,y,z directions

Note:
1. Node numbering for this element is different from that for the eight-node brick (Brick8N) element.
2. The valid queries to a Quad element when creating an ElementRecorder object are ‘forces’, ‘stresses’, ‘strains’,
and ‘material $matNum matArg1 matArg2 . . . ’ Where $matNum refers to the material object at the integration
point corresponding to the node numbers in the isoparametric domain.

See also:
Notes

80 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

Twenty Node Brick Element

The element is used to construct a twenty-node three dimensional element object

element(’20NodeBrick’, eleTag, *eleNodes, matTag, bf1, bf2, bf3, massDen)

eleTag (int) unique element object tag

eleNodes (list (int)) a list of twenty element nodes, input order is shown in notes below
matTag (int) material tag associated with previsouly-defined NDMaterial object
bf1 bf2 bf3 (float) body force in the direction of global coordinates x, y and z
massDen (float) mass density (mass/volume)

Note: The valid queries to a 20NodeBrick element when creating an ElementRecorder object are ‘force,’ ‘stiffness,’
stress’, ‘gausspoint’ or ‘plastic’. The output is given as follows:
1. ‘stress’
the six stress components from each Gauss points are output by the order: sigma_xx, sigma_yy, sigma_zz,
sigma_xy, sigma_xz,sigma_yz
2. ‘gausspoint’
the coordinates of all Gauss points are printed out
3. ‘plastic’
the equivalent deviatoric plastic strain from each Gauss point is output in the same order as the coordinates are
printed

See also:
Notes

SSPbrick Element

This command is used to construct a SSPbrick element object.

element(’SSPbrick’, eleTag, *eleNodes, matTag, <b1, b2, b3>)

eleTag (int) unique element object tag

eleNodes (list a list of eight element nodes in bottom and top faces and in counter-clockwise order
(int))
matTag (int) unique integer tag associated with previously-defined nDMaterial object
b1 b2 b3 (float) constant body forces in global x-, y-, and z-directions, respectively (optional, default
= 0.0)

The SSPbrick element is an eight-node hexahedral element using physically stabilized single-point integration (SSP –>
Stabilized Single Point). The stabilization incorporates an enhanced assumed strain field, resulting in an element which
is free from volumetric and shear locking. The elimination of shear locking results in greater coarse mesh accuracy in
bending dominated problems, and the elimination of volumetric locking improves accuracy in nearly-incompressible
problems. Analysis times are generally faster than corresponding full integration elements.

1.4. Model Commands 81

OpenSeesPy Documentation, Release 3.5.1.3

Note:
1. Valid queries to the SSPbrick element when creating an ElementalRecorder object correspond to those for the
nDMaterial object assigned to the element (e.g., ‘stress’, ‘strain’). Material response is recorded at the single
integration point located in the center of the element.
2. The SSPbrick element was designed with intentions of duplicating the functionality of the stdBrick Element. If
an example is found where the SSPbrick element cannot do something that works for the stdBrick Element, e.g.,
material updating, please contact the developers listed below so the bug can be fixed.

See also:
Notes

Tetrahedron Elements

1. FourNodeTetrahedron

FourNodeTetrahedron

This command is used to construct a standard four-node tetrahedron element objec with one-point Gauss integration.
element(’FourNodeTetrahedron’, eleTag, *eleNodes, matTag, <b1, b2, b3>)

eleTag (int) unique element object tag

eleNodes (list (int)) a list of four element nodes
matTag (int) tag of nDMaterial
b1 b2 b3 (float) body forces in global x,y,z directions

See also:
Notes

UC San Diego u-p element (saturated soil)

1. Four Node Quad u-p Element

2. Brick u-p Element
3. BbarQuad u-p Element
4. BbarBrick u-p Element
5. Nine Four Node Quad u-p Element
6. Twenty Eight Node Brick u-p Element

Four Node Quad u-p Element

FourNodeQuadUP is a four-node plane-strain element using bilinear isoparametric formulation. This element is im-
plemented for simulating dynamic response of solid-fluid fully coupled material, based on Biot’s theory of porous
medium. Each element node has 3 degrees-of-freedom (DOF): DOF 1 and 2 for solid displacement (u) and DOF 3 for
fluid pressure (p).

82 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

element(’quadUP’, eleTag, *eleNodes, thick, matTag, bulk, fmass, hPerm, vPerm, <b1=0, b2=0, t=0>)

eleTag unique element object tag

(int)
eleNodesa list of four element nodes in counter-clockwise order
(list
(int))
thick Element thickness
(float)
matTag Tag of an NDMaterial object (previously defined) of which the element is composed
(int)
bulk Combined undrained bulk modulus Bc relating changes in pore pressure and volumetric strain,
(float) may be approximated by: 𝐵𝑐 ≈ 𝐵𝑓 /𝑛
where 𝐵𝑓 is the bulk modulus of fluid phase (2.2 × 106 kPa (or 3.191 × 105 psi) for water), and
n the initial porosity.
fmass Fluid mass density
(float)
hPerm, Permeability coefficient in horizontal and vertical directions respectively.
vPerm
(float)
b1, b2 Optional gravity acceleration components in horizontal and vertical directions respectively (de-
(float) faults are 0.0)
t (float) Optional uniform element normal traction, positive in tension (default is 0.0)

See also:
Notes

Brick u-p Element

BrickUP is an 8-node hexahedral linear isoparametric element. Each node has 4 degrees-of-freedom (DOF): DOFs 1
to 3 for solid displacement (u) and DOF 4 for fluid pressure (p). This element is implemented for simulating dynamic
response of solid-fluid fully coupled material, based on Biot’s theory of porous medium.
element(’brickUP’, eleTag, *eleNodes, matTag, bulk, fmass, permX, permY, permZ, <bX=0, bY=0, bZ=0>)

1.4. Model Commands 83

OpenSeesPy Documentation, Release 3.5.1.3

eleTag unique element object tag

(int)
eleNodes a list of eight element nodes
(list (int))
matTag Tag of an NDMaterial object (previously defined) of which the element is composed
(int)
bulk Combined undrained bulk modulus Bc relating changes in pore pressure and volumetric strain,
(float) may be approximated by: 𝐵𝑐 ≈ 𝐵𝑓 /𝑛
where 𝐵𝑓 is the bulk modulus of fluid phase (2.2 × 106 kPa (or 3.191 × 105 psi) for water),
and n the initial porosity.
fmass Fluid mass density
(float)
permX, Permeability coefficients in x, y, and z directions respectively.
permY,
permZ
(float)
bX, bY, bZ Optional gravity acceleration components in x, y, and z directions directions respectively (de-
(float) faults are 0.0)

See also:
Notes

BbarQuad u-p Element

bbarQuadUP is a four-node plane-strain mixed volume/pressure element, which uses a tri-linear isoparametric for-
mulation. This element is implemented for simulating dynamic response of solid-fluid fully coupled material, based
on Biot’s theory of porous medium. Each element node has 3 degrees-of-freedom (DOF): DOF 1 and 2 for solid
displacement (u) and DOF 3 for fluid pressure (p).
element(’bbarQuadUP’, eleTag, *eleNodes, thick, matTag, bulk, fmass, hPerm, vPerm, <b1=0, b2=0, t=0>)

84 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

eleTag unique element object tag

(int)
eleNodesa list of four element nodes in counter-clockwise order
(list
(int))
thick Element thickness
(float)
matTag Tag of an NDMaterial object (previously defined) of which the element is composed
(int)
bulk Combined undrained bulk modulus Bc relating changes in pore pressure and volumetric strain,
(float) may be approximated by: 𝐵𝑐 ≈ 𝐵𝑓 /𝑛
where 𝐵𝑓 is the bulk modulus of fluid phase (2.2 × 106 kPa (or 3.191 × 105 psi) for water), and
n the initial porosity.
fmass Fluid mass density
(float)
hPerm, Permeability coefficient in horizontal and vertical directions respectively.
vPerm
(float)
b1, b2 Optional gravity acceleration components in horizontal and vertical directions respectively (de-
(float) faults are 0.0)
t (float) Optional uniform element normal traction, positive in tension (default is 0.0)

See also:
Notes

BbarBrick u-p Element

bbarBrickUP is a 8-node mixed volume/pressure element, which uses a tri-linear isoparametric formulation.
Each node has 4 degrees-of-freedom (DOF): DOFs 1 to 3 for solid displacement (u) and DOF 4 for fluid pressure (p).
This element is implemented for simulating dynamic response of solid-fluid fully coupled material, based on Biot’s
theory of porous medium.
element(’bbarBrickUP’, eleTag, *eleNodes, matTag, bulk, fmass, permX, permY, permZ, <bX=0, bY=0,
bZ=0>)

1.4. Model Commands 85

OpenSeesPy Documentation, Release 3.5.1.3

eleTag unique element object tag

(int)
eleNodes a list of eight element nodes
(list (int))
matTag Tag of an NDMaterial object (previously defined) of which the element is composed
(int)
bulk Combined undrained bulk modulus Bc relating changes in pore pressure and volumetric strain,
(float) may be approximated by: 𝐵𝑐 ≈ 𝐵𝑓 /𝑛
where 𝐵𝑓 is the bulk modulus of fluid phase (2.2 × 106 kPa (or 3.191 × 105 psi) for water),
and n the initial porosity.
fmass Fluid mass density
(float)
permX, Permeability coefficients in x, y, and z directions respectively.
permY,
permZ
(float)
bX, bY, bZ Optional gravity acceleration components in x, y, and z directions directions respectively (de-
(float) faults are 0.0)

See also:
Notes

Nine Four Node Quad u-p Element

Nine_Four_Node_QuadUP is a 9-node quadrilateral plane-strain element. The four corner nodes have 3 degrees-of-
freedom (DOF) each: DOF 1 and 2 for solid displacement (u) and DOF 3 for fluid pressure (p). The other five nodes
have 2 DOFs each for solid displacement. This element is implemented for simulating dynamic response of solid-fluid
fully coupled material, based on Biot’s theory of porous medium.
element(’9_4_QuadUP’, eleTag, *eleNodes, thick, matTag, bulk, fmass, hPerm, vPerm, <b1=0, b2=0>)

eleTag unique element object tag

(int)
eleNodesa list of nine element nodes
(list
(int))
thick Element thickness
(float)
matTag Tag of an NDMaterial object (previously defined) of which the element is composed
(int)
bulk Combined undrained bulk modulus Bc relating changes in pore pressure and volumetric strain,
(float) may be approximated by: 𝐵𝑐 ≈ 𝐵𝑓 /𝑛
where 𝐵𝑓 is the bulk modulus of fluid phase (2.2 × 106 kPa (or 3.191 × 105 psi) for water), and
n the initial porosity.
fmass Fluid mass density
(float)
hPerm, Permeability coefficient in horizontal and vertical directions respectively.
vPerm
(float)
b1, b2 Optional gravity acceleration components in horizontal and vertical directions respectively (de-
(float) faults are 0.0)

86 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

See also:
Notes

Twenty Eight Node Brick u-p Element

Twenty_Eight_Node_BrickUP is a 20-node hexahedral isoparametric element.

The eight corner nodes have 4 degrees-of-freedom (DOF) each: DOFs 1 to 3 for solid displacement (u) and DOF 4
for fluid pressure (p). The other nodes have 3 DOFs each for solid displacement. This element is implemented for
simulating dynamic response of solid-fluid fully coupled material, based on Biot’s theory of porous medium.
element(’20_8_BrickUP’, eleTag, *eleNodes, matTag, bulk, fmass, permX, permY, permZ, <bX=0, bY=0,
bZ=0>)

eleTag unique element object tag

(int)
eleNodes a list of twenty element nodes
(list (int))
matTag Tag of an NDMaterial object (previously defined) of which the element is composed
(int)
bulk Combined undrained bulk modulus Bc relating changes in pore pressure and volumetric strain,
(float) may be approximated by: 𝐵𝑐 ≈ 𝐵𝑓 /𝑛
where 𝐵𝑓 is the bulk modulus of fluid phase (2.2 × 106 kPa (or 3.191 × 105 psi) for water),
and n the initial porosity.
fmass Fluid mass density
(float)
permX, Permeability coefficients in x, y, and z directions respectively.
permY,
permZ
(float)
bX, bY, bZ Optional gravity acceleration components in x, y, and z directions directions respectively (de-
(float) faults are 0.0)

See also:
Notes

Other u-p elements

1. SSPquadUP Element
2. SSPbrickUP Element

SSPquadUP Element

This command is used to construct a SSPquadUP element object.

element(’SSPquadUP’, eleTag, *eleNodes, matTag, thick, fBulk, fDen, k1, k2, void, alpha, <b1=0.0,
b2=0.0>)

1.4. Model Commands 87

OpenSeesPy Documentation, Release 3.5.1.3

eleTag (int) unique element object tag

eleNodes (list a list of four element nodes in counter-clockwise order
(int))
matTag (int) unique integer tag associated with previously-defined nDMaterial object
thick (float) thickness of the element in out-of-plane direction
fBulk (float) bulk modulus of the pore fluid
fDen (float) mass density of the pore fluid
k1 k2 (float) permeability coefficients in global x- and y-directions, respectively
void (float) voids ratio
alpha (float) spatial pressure field stabilization parameter (see discussion below for more informa-
tion)
b1 b2 (float) constant body forces in global x- and y-directions, respectively (optional, default =
0.0) - See Note 3

The SSPquadUP element is an extension of the SSPquad Element for use in dynamic plane strain
analysis of fluid saturated porous media. A mixed displacement-pressure (u-p) formulation is used,
based upon the work of Biot as extended by Zienkiewicz and Shiomi (1984).

The physical stabilization necessary to allow for reduced integration incorporates an assumed strain field in which the
volumetric dilation and the shear strain associated with the the hourglass modes are zero, resulting in an element which
is free from volumetric and shear locking. The elimination of shear locking results in greater coarse mesh accuracy in
bending dominated problems, and the elimination of volumetric locking improves accuracy in nearly-incompressible
problems. Analysis times are generally faster than corresponding full integration elements.
Equal-order interpolation is used for the displacement and pressure fields, thus, the SSPquadUP element does not
inherently pass the inf-sup condition, and is not fully acceptable in the incompressible-impermeable limit (the QuadUP
Element has the same issue). A stabilizing parameter is employed to permit the use of equal-order interpolation for
the SSPquadUP element. This parameter $alpha can be computed as

𝛼 = 0.25 * (ℎ2 )/(𝑑𝑒𝑛 * 𝑐2 )

where h is the element size, c is the speed of elastic wave propagation in the solid phase, and den is the mass density
of the solid phase. The $alpha parameter should be a small number. With a properly defined $alpha parameter, the
SSPquadUP element can produce comparable results to a higher-order element such as the 9_4_QuadUP Element at a
significantly lower computational cost and with a greater ease in mesh generation.
The full formulation for the SSPquadUP element can be found in McGann et al. (2012) along with several example
applications.

Note:
1. The SSPquadUP element will only work in dynamic analysis.
2. For saturated soils, the mass density input into the associated nDMaterial object should be the saturated mass
density.
3. When modeling soil, the body forces input into the SSPquadUP element should be the components of the
gravitational vector, not the unit weight.
4. Fixing the pore pressure degree-of-freedom (dof 3) at a node is a drainage boundary condition at which zero
pore pressure will be maintained throughout the analysis. Leaving the third dof free allows pore pressures to
build at that node.
5. Valid queries to the SSPquadUP element when creating an ElementalRecorder object correspond to those for
the nDMaterial object assigned to the element (e.g., ‘stress’, ‘strain’). Material response is recorded at the single
integration point located in the center of the element.

88 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

6. The SSPquadUP element was designed with intentions of duplicating the functionality of the QuadUP Element.
If an example is found where the SSPquadUP element cannot do something that works for the QuadUP Element,
e.g., material updating, please contact the developers listed below so the bug can be fixed.

See also:
Notes

SSPbrickUP Element

This command is used to construct a SSPbrickUP element object.

element(’SSPbrickUP’, eleTag, *eleNodes, matTag, fBulk, fDen, k1, k2, k3, void, alpha, <b1, b2, b3>)

eleTag (int) unique element object tag

eleNodes (list a list of eight element nodes in counter-clockwise order
(int))
matTag (float) unique integer tag associated with previously-defined nDMaterial object
fBulk (float) bulk modulus of the pore fluid
fDen (float) mass density of the pore fluid
k1 k2 k3 (float) permeability coefficients in global x-, y-, and z-directions, respectively
void (float) voids ratio
alpha (float) spatial pressure field stabilization parameter (see discussion below for more informa-
tion)
b1 b2 b3 (float) constant body forces in global x-, y-, and z-directions, respectively (optional, default =
0.0) - See Note 3

The SSPbrickUP element is an extension of the SSPbrick Element for use in dynamic 3D analysis of fluid saturated
porous media. A mixed displacement-pressure (u-p) formulation is used, based upon the work of Biot as extended by
Zienkiewicz and Shiomi (1984).
The physical stabilization necessary to allow for reduced integration incorporates an enhanced assumed strain field,
resulting in an element which is free from volumetric and shear locking. The elimination of shear locking results in
greater coarse mesh accuracy in bending dominated problems, and the elimination of volumetric locking improves
accuracy in nearly-incompressible problems. Analysis times are generally faster than corresponding full integration
elements.
Equal-order interpolation is used for the displacement and pressure fields, thus, the SSPbrickUP element does not
inherently pass the inf-sup condition, and is not fully acceptable in the incompressible-impermeable limit (the brickUP
Element has the same issue). A stabilizing parameter is employed to permit the use of equal-order interpolation for
the SSPbrickUP element. This parameter $alpha can be computed as

𝛼 = ℎ2 /(4 * (𝐾𝑠 + (4/3) * 𝐺𝑠 ))

where ℎ is the element size, and 𝐾𝑠 and 𝐺𝑠 are the bulk and shear moduli for the solid phase. The 𝛼 parameter should
be a small number. With a properly defined 𝛼 parameter, the SSPbrickUP element can produce comparable results
to a higher-order element such as the 20_8_BrickUP Element at a significantly lower computational cost and with a
greater ease in mesh generation.

Note:
1. The SSPbrickUP element will only work in dynamic analysis.

1.4. Model Commands 89

OpenSeesPy Documentation, Release 3.5.1.3

2. For saturated soils, the mass density input into the associated nDMaterial object should be the saturated mass
density.
3. When modeling soil, the body forces input into the SSPbrickUP element should be the components of the
gravitational vector, not the unit weight.
4. Fixing the pore pressure degree-of-freedom (dof 4) at a node is a drainage boundary condition at which zero
pore pressure will be maintained throughout the analysis. Leaving the fourth dof free allows pore pressures to
build at that node.
5. Valid queries to the SSPbrickUP element when creating an ElementalRecorder object correspond to those for
the nDMaterial object assigned to the element (e.g., ‘stress’, ‘strain’). Material response is recorded at the single
integration point located in the center of the element.
6. The SSPbrickUP element was designed with intentions of duplicating the functionality of the brickUP Element.
If an example is found where the SSPbrickUP element cannot do something that works for the brickUP Element,
e.g., material updating, please contact the developers listed below so the bug can be fixed.

See also:
Notes

Contact Elements

1. SimpleContact2D
2. SimpleContact3D
3. BeamContact2D
4. BeamContact3D
5. BeamEndContact3D

SimpleContact2D

This command is used to construct a SimpleContact2D element object.

element(’SimpleContact2D’, eleTag, iNode, jNode, cNode, lNode, matTag, gTol, fTol)

eleTag (int) unique element object tag

iNode jNode (int) retained nodes (-ndm 2 -ndf 2)
cNode (int) constrained node (-ndm 2 -ndf 2)
lNode (int) Lagrange multiplier node (-ndm 2 -ndf 2)
matTag (int) unique integer tag associated with previously-defined nDMaterial object
gTol (float) gap tolerance
fTol (float) force tolerance

The SimpleContact2D element is a two-dimensional node-to-segment contact element which defines a frictional con-
tact interface between two separate bodies. The master nodes are the nodes which define the endpoints of a line
segment on the first body, and the slave node is a node from the second body. The Lagrange multiplier node is
required to enforce the contact condition. This node should not be shared with any other element in the domain.
Information on the theory behind this element can be found in, e.g. Wriggers (2002).

Note:

90 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

1. The SimpleContact2D element has been written to work exclusively with the ContactMaterial2D nDMaterial
object.
2. The valid recorder queries for this element are:
1. force - returns the contact force acting on the slave node in vector form.
2. frictionforce - returns the frictional force acting on the slave node in vector form.
3. forcescalar - returns the scalar magnitudes of the normal and tangential contact forces.
4. The SimpleContact2D elements are set to consider frictional behavior as a default, but the frictional state
of the SimpleContact2D element can be changed from the input file using the setParameter command.
When updating, value of 0 corresponds to the frictionless condition, and a value of 1 signifies the inclusion
of friction. An example command for this update procedure is provided below
3. The SimpleContact2D element works well in static and pseudo-static analysis situations.
4. In transient analysis, the presence of the contact constraints can effect the stability of commonly-used time
integration methods in the HHT or Newmark family (e.g., Laursen, 2002). For this reason, use of alternative
time-integration methods which numerically damp spurious high frequency behavior may be required. The
TRBDF2 integrator is an effective method for this purpose. The Newmark integrator can also be effective with
proper selection of the gamma and beta coefficients. The trapezoidal rule, i.e., Newmark with gamma = 0.5 and
beta = 0.25, is particularly prone to instability related to the contact constraints and is not recommended.

See also:
Notes

SimpleContact3D

This command is used to construct a SimpleContact3D element object.

element(’SimpleContact3D’, eleTag, iNode, jNode, kNode, lNode, cNode, lagr_node, matTag, gTol, fTol)

eleTag (int) unique element object tag

iNode jNode kNode lNode master nodes (-ndm 3 -ndf 3)
(int)
cNode (int) constrained node (-ndm 3 -ndf 3)
lagr_node (int) Lagrange multiplier node (-ndm 3 -ndf 3)
matTag (int) unique integer tag associated with previously-defined nDMaterial ob-
ject
gTol (float) gap tolerance
fTol (float) force tolerance

The SimpleContact3D element is a three-dimensional node-to-surface contact element which defines a frictional con-
tact interface between two separate bodies. The master nodes are the nodes which define a surface of a hexahedral
element on the first body, and the slave node is a node from the second body. The Lagrange multiplier node is required
to enforce the contact condition. This node should not be shared with any other element in the domain. Information
on the theory behind this element can be found in, e.g. Wriggers (2002).

Note:
1. The SimpleContact3D element has been written to work exclusively with the ContactMaterial3D nDMaterial
object.

1.4. Model Commands 91

OpenSeesPy Documentation, Release 3.5.1.3

2. The valid recorder queries for this element are:

1. force - returns the contact force acting on the slave node in vector form.
2. frictionforce - returns the frictional force acting on the slave node in vector form.
3. forcescalar - returns the scalar magnitudes of the single normal and two tangential contact forces.
4. The SimpleContact3D elements are set to consider frictional behavior as a default, but the frictional state
of the SimpleContact3D element can be changed from the input file using the setParameter command.
When updating, value of 0 corresponds to the frictionless condition, and a value of 1 signifies the inclusion
of friction. An example command for this update procedure is provided below
3. The SimpleContact3D element works well in static and pseudo-static analysis situations.
4. In transient analysis, the presence of the contact constraints can effect the stability of commonly-used time
integration methods in the HHT or Newmark family (e.g., Laursen, 2002). For this reason, use of alternative
time-integration methods which numerically damp spurious high frequency behavior may be required. The
TRBDF2 integrator is an effective method for this purpose. The Newmark integrator can also be effective with
proper selection of the gamma and beta coefficients. The trapezoidal rule, i.e., Newmark with gamma = 0.5 and
beta = 0.25, is particularly prone to instability related to the contact constraints and is not recommended.

See also:
Notes

BeamContact2D

This command is used to construct a BeamContact2D element object.

element(’BeamContact2D’, eleTag, iNode, jNode, sNode, lNode, matTag, width, gTol, fTol, <cFlag>)

eleTag unique element object tag

(int)
iNode master nodes (-ndm 2 -ndf 3)
jNode
(int)
sNode slave node (-ndm 2 -ndf 2)
(int)
lNode Lagrange multiplier node (-ndm 2 -ndf 2)
(int)
matTag unique integer tag associated with previously-defined nDMaterial object
(int)
width the width of the wall represented by the beam element in plane strain
(float)
gTol gap tolerance
(float)
fTol force tolerance
(float)
cFlag optional initial contact flag
(int) cFlag = 0 >> contact between bodies is initially assumed (DEFAULT)
cFlag = 1 >> no contact between bodies is initially assumed

The BeamContact2D element is a two-dimensional beam-to-node contact element which defines a frictional contact
interface between a beam element and a separate body. The master nodes (3 DOF) are the endpoints of the beam

92 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

element, and the slave node (2 DOF) is a node from a second body. The Lagrange multiplier node (2 DOF) is required
to enforce the contact condition. Each contact element should have a unique Lagrange multiplier node. The Lagrange
multiplier node should not be fixed, otherwise the contact condition will not work.
Under plane strain conditions in 2D, a beam element represents a unit thickness of a wall. The width is the dimension
of this wall in the 2D plane. This width should be built-in to the model to ensure proper enforcement of the con-
tact condition. The Excavation Supported by Cantilevered Sheet Pile Wall practical example provides some further
examples and discussion on the usage of this element.

Note:
1. The BeamContact2D element has been written to work exclusively with the ContactMaterial2D nDMaterial
object.
2. The valid recorder queries for this element are:
1. force - returns the contact force acting on the slave node in vector form.
2. frictionforce - returns the frictional force acting on the slave node in vector form.
3. forcescalar - returns the scalar magnitudes of the normal and tangential contact forces.
4. masterforce - returns the reactions (forces and moments) acting on the master nodes.
5. The BeamContact2D elements are set to consider frictional behavior as a default, but the frictional state of
the BeamContact2D element can be changed from the input file using the setParameter command. When
updating, value of 0 corresponds to the frictionless condition, and a value of 1 signifies the inclusion of
friction. An example command for this update procedure is provided below
3. The BeamContact2D element works well in static and pseudo-static analysis situations.
4. In transient analysis, the presence of the contact constraints can effect the stability of commonly-used time
integration methods in the HHT or Newmark family (e.g., Laursen, 2002). For this reason, use of alternative
time-integration methods which numerically damp spurious high frequency behavior may be required. The
TRBDF2 integrator is an effective method for this purpose. The Newmark integrator can also be effective with
proper selection of the gamma and beta coefficients. The trapezoidal rule, i.e., Newmark with gamma = 0.5 and
beta = 0.25, is particularly prone to instability related to the contact constraints and is not recommended.

See also:
Notes

BeamContact3D

This command is used to construct a BeamContact3D element object.

element(’BeamContact3D’, eleTag, iNode, jNode, cNode, lNode, radius, crdTransf, matTag, gTol, fTol,
<cFlag>)

1.4. Model Commands 93

OpenSeesPy Documentation, Release 3.5.1.3

eleTag unique element object tag

(int)
iNode master nodes (-ndm 3 -ndf 6)
jNode
(int)
cNode constrained node (-ndm 3 -ndf 3)
(int)
lNode Lagrange multiplier node (-ndm 3 -ndf 3)
(int)
radius constant radius of circular beam associated with beam element
(float)
crdTransf unique integer tag associated with previously-defined geometricTransf object
(int)
matTag unique integer tag associated with previously-defined nDMaterial object
(int)
gTol gap tolerance
(float)
fTol force tolerance
(float)
cFlag optional initial contact flag
(int) cFlag = 0 >> contact between bodies is initially assumed (DEFAULT)
cFlag = 1 >> no contact between bodies is initially assumed

The BeamContact3D element is a three-dimensional beam-to-node contact element which defines a frictional contact
interface between a beam element and a separate body. The master nodes (6 DOF) are the endpoints of the beam
element, and the slave node (3 DOF) is a node from a second body. The Lagrange multiplier node (3 DOF) is required
to enforce the contact condition. Each contact element should have a unique Lagrange multiplier node. The Lagrange
multiplier node should not be fixed, otherwise the contact condition will not work.

Note:
1. The BeamContact3D element has been written to work exclusively with the ContactMaterial3D nDMaterial
object.
2. The valid recorder queries for this element are:
1. force - returns the contact force acting on the slave node in vector form.
2. frictionforce - returns the frictional force acting on the slave node in vector form.
3. forcescalar - returns the scalar magnitudes of the single normal and two tangential contact forces.
4. masterforce - returns the reactions (forces only) acting on the master nodes.
5. mastermoment - returns the reactions (moments only) acting on the master nodes.
6. masterreaction - returns the full reactions (forces and moments) acting on the master nodes.
7. The BeamContact3D elements are set to consider frictional behavior as a default, but the frictional state of
the BeamContact3D element can be changed from the input file using the setParameter command. When
updating, value of 0 corresponds to the frictionless condition, and a value of 1 signifies the inclusion of
friction. An example command for this update procedure is provided below
3. The BeamContact3D element works well in static and pseudo-static analysis situations.
4. In transient analysis, the presence of the contact constraints can effect the stability of commonly-used time
integration methods in the HHT or Newmark family (e.g., Laursen, 2002). For this reason, use of alternative
time-integration methods which numerically damp spurious high frequency behavior may be required. The

94 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

TRBDF2 integrator is an effective method for this purpose. The Newmark integrator can also be effective with
proper selection of the gamma and beta coefficients. The trapezoidal rule, i.e., Newmark with gamma = 0.5 and
beta = 0.25, is particularly prone to instability related to the contact constraints and is not recommended.

See also:
Notes

BeamEndContact3D

This command is used to construct a BeamEndContact3D element object.

element(’BeamEndContact3D’, eleTag, iNode, jNode, cNode, lNode, radius, gTol, fTol, <cFlag>)

eleTag unique element object tag

(int)
iNode master node from the beam (-ndm 3 -ndf 6)
(int)
jNode the remaining node on the beam element with iNode (-ndm 3 -ndf 6)
(int)
cNode constrained node (-ndm 3 -ndf 3)
(int)
lNode Lagrange multiplier node (-ndm 3 -ndf 3)
(int)
radius radius of circular beam associated with beam element
(float)
gTol gap tolerance
(float)
fTol force tolerance
(float)
cFlag optional initial contact flag
(float) cFlag = 0 >> contact between bodies is initially assumed (DEFAULT) cFlag1 = 1 >> no
contact between bodies is initially assumed

The BeamEndContact3D element is a node-to-surface contact element which defines a normal contact interface be-
tween the end of a beam element and a separate body. The first master node ($iNode) is the beam node which is at the
end of the beam (i.e. only connected to a single beam element), the second node ($jNode) is the remaining node on
the beam element in question. The slave node is a node from a second body. The Lagrange multiplier node is required
to enforce the contact condition. This node should not be shared with any other element in the domain, and should be
created with the same number of DOF as the slave node.
The BeamEndContact3D element enforces a contact condition between a fictitious circular plane associated with a
beam element and a node from a second body. The normal direction of the contact plane coincides with the endpoint
tangent of the beam element at the master beam node ($iNode). The extents of this circular plane are defined by
the radius input parameter. The master beam node can only come into contact with a slave node which is within the
extents of the contact plane. There is a lag step associated with changing between the ‘in contact’ and ‘not in contact’
conditions.
This element was developed for use in establishing a contact condition for the tip of a pile modeled as using beam
elements and the underlying soil elements in three-dimensional analysis.

Note:

1.4. Model Commands 95

OpenSeesPy Documentation, Release 3.5.1.3

1. The BeamEndContact3D element does not use a material object.

2. The valid recorder queries for this element are:
1. force - returns the contact force acting on the slave node in vector form.
2. masterforce - returns the reactions (forces and moments) acting on the master node.
3. The BeamEndContact3D element works well in static and pseudo-static analysis situations.
3. In transient analysis, the presence of the contact constraints can effect the stability of commonly-used time
integration methods in the HHT or Newmark family (e.g., Laursen, 2002). For this reason, use of alternative
time-integration methods which numerically damp spurious high frequency behavior may be required. The
TRBDF2 integrator is an effective method for this purpose. The Newmark integrator can also be effective with
proper selection of the gamma and beta coefficients. The trapezoidal rule, i.e., Newmark with gamma = 0.5 and
beta = 0.25, is particularly prone to instability related to the contact constraints and is not recommended.

See also:
Notes

Cable Elements

1. CatenaryCableElement

CatenaryCableElement

This command is used to construct a catenary cable element object.

element(’CatenaryCable’, eleTag, iNode, jNode, weight, E, A, L0, alpha, temperature_change, rho, errorTol,
Nsubsteps, massType)

eleTag (int) unique element object tag

iNode jNode (int) end nodes (3 dof per node)
weight (float) undefined
E (float) elastic modulus of the cable material
A (float) cross-sectional area of element
L0 (float) unstretched length of the cable
alpha (float) coefficient of thermal expansion
temperature_changetemperature change for the element
(float)
rho (float) mass per unit length
errorTol (float) allowed tolerance for within-element equilbrium (Newton-Rhapson iterations)
Nsubsteps (int) number of within-element substeps into which equilibrium iterations are subdi-
vided (not number of steps to convergence)
massType (int) Mass matrix model to use (massType = 0 lumped mass matrix, massType = 1
rigid-body mass matrix (in development))

This cable is a flexibility-based formulation of the catenary cable. An iterative scheme is used internally to compute
equilibrium. At each iteration, node i is considered fixed while node j is free. End-forces are applied at node-j and
its displacements computed. Corrections to these forces are applied iteratively using a Newton-Rhapson scheme (with
optional sub-stepping via $Nsubsteps) until nodal displacements are within the provided tolerance ($errortol). When
convergence is reached, a stiffness matrix is computed by inversion of the flexibility matrix and rigid-body mode
injection.

96 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

Note:
1. The stiffness of the cable comes from the large-deformation interaction between loading and cable shape. There-
fore, all cables must have distributed forces applied to them. See example. Should not work for only nodal
forces.
2. Valid queries to the CatenaryCable element when creating an ElementalRecorder object correspond to ‘forces’,
which output the end-forces of the element in global coordinates (3 for each node).
3. Only the lumped-mass formulation is currently available.
4. The element does up 100 internal iterations. If convergence is not achieved, will result in error and some
diagnostic information is printed out.

See also:
Notes

PFEM Elements

1. PFEMElementBubble
2. PFEMElementCompressible

PFEMElementBubble

element(’PFEMElementBubble’, eleTag, *eleNodes, rho, mu, b1, b2, <b3>, <thickness, kappa>)
Create a PFEM Bubble element, which is a fluid element for FSI analysis.

eleTag (int) tag of the element

eleNodes (list (int)) A list of three or four element nodes, four are required for 3D
nd4 (int) tag of node 4 (required for 3D)
rho (float) fluid density
mu (float) fluid viscosity
b1 (float) body body acceleration in x direction
b2 (float) body body acceleration in y direction
b3 (float) body body acceleration in z direction (required for 3D)
thickness (float) element thickness (required for 2D)
kappa (float) fluid bulk modulus (optional)

PFEMElementCompressible

element(’PFEMElementCompressible’, eleTag, *eleNodes, rho, mu, b1, b2, <thickness, kappa>)

Create a PFEM compressible element, which is a fluid element for FSI analysis.

1.4. Model Commands 97

OpenSeesPy Documentation, Release 3.5.1.3

eleTag (int) tag of the element

eleNodes (list (int)) A list of four element nodes, last one is middle node
rho (float) fluid density
mu (float) fluid viscosity
b1 (float) body body acceleration in x direction
b2 (float) body body acceleration in y direction
thickness (float) element thickness (optional)
kappa (float) fluid bulk modulus (optional)

Misc.

1. SurfaceLoad Element
2. VS3D4
3. AC3D8
4. ASI3D8
5. AV3D4

SurfaceLoad Element

This command is used to construct a SurfaceLoad element object.

element(’SurfaceLoad’, eleTag, *eleNodes, p)

eleTag (int) unique element object tag

eleNodes (list the four nodes defining the element, input in counterclockwise order (-ndm 3 -ndf
(int)) 3)
p (float) applied pressure loading normal to the surface, outward is positive, inward is nega-
tive

The SurfaceLoad element is a four-node element which can be used to apply surface pressure loading to 3D brick
elements. The SurfaceLoad element applies energetically-conjugate forces corresponding to the input scalar pressure
to the nodes associated with the element. As these nodes are shared with a 3D brick element, the appropriate nodal
loads are therefore applied to the brick.

Note:
1. There are no valid ElementalRecorder queries for the SurfaceLoad element. Its sole purpose is to apply nodal
forces to the adjacent brick element.
2. The pressure loading from the SurfaceLoad element can be applied in a load pattern. See the analysis example
below.

See also:
Notes

98 Chapter 1. Developer
 OpenSeesPy Documentation, Release 3.5.1.3

VS3D4

This command is used to construct a four-node 3D viscous-spring boundary quad element object based on a bilinear
isoparametric formulation.
element(’VS3D4’, eleTag, *eleNodes, E, G, rho, R, alphaN, alphaT)

eleTag (int) unique element object tag

eleNodes (list (int)) 4 end nodes
E (float) Young’s Modulus of element material
G (float) Shear Modulus of element material
rho (float) Mass Density of element material
R (float) distance from the scattered wave source to the boundary
alphaN (float) correction parameter in the normal direction
alphaT (float) correction parameter in the tangential direction

Note: Reference: Liu J, Du Y, Du X, et al. 3D viscous-spring artificial boundary in time domain. Earthquake
Engineering and Engineering Vibration, 2006, 5(1):93-102

See also:
Notes

AC3D8

This command is used to construct an eight-node 3D brick acoustic element object based on a trilinear isoparametric
formulation.
element(’AC3D8’, eleTag, *eleNodes, matTag)

eleTag (int) unique element object tag

eleNodes (list (int)) 8 end nodes
matTag (int) Material Tag of previously defined nD material

Note: Reference: ABAQUS theory manual. (2.9.1 Coupled acoustic-structural medium analysis)

See also:
Notes

ASI3D8

This command is used to construct an eight-node zero-thickness 3D brick acoustic-structure interface element object
based on a bilinear isoparametric formulation. The nodes in the acoustic domain share the same coordinates with the
nodes in the solid domain.

1.4. Model Commands 99

OpenSeesPy Documentation, Release 3.5.1.3

element(’ASI3D8’, eleTag, *eleNodes1, *eleNodes2)

eleTag (int) unique element object tag

*eleNodes1 (list (int)) four nodes defining structure domain of element boundaries
*eleNodes2 (list (int)) four nodes defining acoustic domain of element boundaries

Note: Reference: ABAQUS theory manual. (2.9.1 Coupled acoustic-structural medium analysis)

See also:
Notes

AV3D4

This command is used to construct a four-node 3D acoustic viscous boundary quad element object based on a bilinear
isoparametric formulation.
element(’AV3D4’, eleTag, *eleNodes, matTag)

eleTag (int) unique element object tag

eleNodes (list (int)) 4 end nodes
matTag (int) Material Tag of previously defined nD material

See also:
Notes

1.4.3 node command

node(nodeTag, *crds, ’-ndf’, ndf, ’-mass’, *mass, ’-disp’, *disp, ’-vel’, *vel, ’-accel’, *accel)
Create a OpenSees node.

nodeTag (int) node tag.

crds (list (float)) nodal coordinates.
ndf (float) nodal ndf. (optional)
mass (list (float)) nodal mass. (optional)
vel (list (float)) nodal velocities. (optional)
accel (list (float)) nodal accelerations. (optional)

1.4.4 sp constraint commands

Create constraints for a single dof of a node.

1. fix command
2. fixX command
3. fixY command
4. fixZ command

100 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

fix command

fix(nodeTag, *constrValues)
Create a homogeneous SP constriant.

nodeTag (int) tag of node to be constrained

constrValues (list (int)) a list of constraint values (0 or 1), must be preceded
with *.
• 0 free
• 1 fixed

For example,

# fully fixed
vals = [1,1,1]
fix(nodeTag, *vals)

fixX command

fixX(x, *constrValues, ’-tol’, tol=1e-10)

Create homogeneous SP constriants.

x (float) x-coordinate of nodes to be constrained

constrValues (list (int)) a list of constraint values (0 or 1), must be preceded
with *.
• 0 free
• 1 fixed

tol (float) user-defined tolerance (optional)

fixY command

fixY(y, *constrValues, ’-tol’, tol=1e-10)

Create homogeneous SP constriants.

y (float) y-coordinate of nodes to be constrained

constrValues (list (int)) a list of constraint values (0 or 1), must be preceded
with *.
• 0 free
• 1 fixed

tol (float) user-defined tolerance (optional)

fixZ command

fixZ(z, *constrValues, ’-tol’, tol=1e-10)

Create homogeneous SP constriants.

1.4. Model Commands 101

OpenSeesPy Documentation, Release 3.5.1.3

z (float) z-coordinate of nodes to be constrained

constrValues (list (int)) a list of constraint values (0 or 1), must be preceded
with *.
• 0 free
• 1 fixed

tol (float) user-defined tolerance (optional)

1.4.5 mp constraint commands

Create constraints for multiple dofs of multiple nodes.

1. equalDOF command
2. equalDOF_Mixed command
3. rigidDiaphragm command
4. rigidLink command

equalDOF command

equalDOF(rNodeTag, cNodeTag, *dofs)

Create a multi-point constraint between nodes.

rNodeTag integer tag identifying the retained, or master node.

(int)
cNodeTag integer tag identifying the constrained, or slave node.
(int)
dofs nodal degrees-of-freedom that are constrained at the cNode to be the same as those at the rNode
(list Valid range is from 1 through ndf, the number of nodal degrees-of-freedom.
(int))

equalDOF_Mixed command

equalDOF_Mixed(rNodeTag, cNodeTag, numDOF, *rcdofs)

Create a multi-point constraint between nodes.

rNodeTaginteger tag identifying the retained, or master node.

(int)
cNodeTaginteger tag identifying the constrained, or slave node.
(int)
numDOF number of dofs to be constrained
(int)
rcdofs nodal degrees-of-freedom that are constrained at the cNode to be the same as those at the rNode
(list Valid range is from 1 through ndf, the number of nodal degrees-of-freedom.
(int)) rcdofs = [rdof1, cdof1, rdof2, cdof2, ...]

102 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

rigidDiaphragm command

rigidDiaphragm(perpDirn, rNodeTag, *cNodeTags)

Create a multi-point constraint between nodes. These objects will constrain certain degrees-of-freedom at the
listed secondary nodes to move as if in a rigid plane with the primary (retained) node. To enforce this constraint,
Transformation constraint handler is recommended.

perpDirn (int) direction perpendicular to the rigid plane (i.e. direction 3 corresponds to the 1-2
plane)
rNodeTag (int) integer tag identifying the retained (primary) node
cNodeTags (list integar tags identifying the constrained (secondary) nodes
(int))

rigidLink command

rigidLink(type, rNodeTag, cNodeTag)

Create a multi-point constraint between nodes.

type (str) string-based argument for rigid-link type:

• 'bar': only the translational degree-of-
freedom will be constrained to be exactly the
same as those at the master node
• 'beam': both the translational and rotational
degrees of freedom are constrained.

rNodeTag (int) integer tag identifying the master node

cNodeTag (int) integar tag identifying the slave node

1.4.6 pressureConstraint command

pressureConstraint(nodeTag, pNodeTag)
Create a pressure constraint for incompressible flow.

nodeTag (int) tag of node to be constrained

pNodeTag (int) tag of extra pressure node, which must exist before calling this command

For example,

ops.node(1, 0.0, 0.0)

ops.node(2, 0.0, 0.0, '-ndf', 1)
ops.pressureConstraint(1, 2)

1.4.7 timeSeries commands

timeSeries(tsType, tsTag, *tsArgs)

This command is used to construct a TimeSeries object which represents the relationship between the time in
the domain, 𝑡, and the load factor applied to the loads, 𝜆, in the load pattern with which the TimeSeries object is
associated, i.e. 𝜆 = 𝐹 (𝑡).

1.4. Model Commands 103

OpenSeesPy Documentation, Release 3.5.1.3

tsType (str) time series type.

tsTag (int) time series tag.
tsArgs (list) a list of time series arguments

The following contain information about available tsType:

1. Constant TimeSeries
2. Linear TimeSeries
3. Trigonometric TimeSeries
4. Triangular TimeSeries
5. Rectangular TimeSeries
6. Pulse TimeSeries
7. Path TimeSeries

Constant TimeSeries

timeSeries(’Constant’, tag, ’-factor’, factor=1.0)

This command is used to construct a TimeSeries object in which the load factor applied remains constant and is
independent of the time in the domain, i.e. 𝜆 = 𝑓 (𝑡) = 𝐶.

tag (int) unique tag among TimeSeries objects.

factor (float) the load factor applied (optional)

Linear TimeSeries

timeSeries(’Linear’, tag, ’-factor’, factor=1.0, ’-tStart’, tStart=0.0)

This command is used to construct a TimeSeries object in which the load factor applied is linearly proportional
to the time in the domain, i.e.
𝜆 = 𝑓 (𝑡) = 𝑐𝐹 𝑎𝑐𝑡𝑜𝑟 * (𝑡 − 𝑡𝑆𝑡𝑎𝑟𝑡). (0 if t < tStart)

tag (int) unique tag among TimeSeries objects

factor (float) Linear factor (optional)
tStart (float) start time (optional)

Trigonometric TimeSeries

timeSeries(’Trig’, tag, tStart, tEnd, period, ’-factor’, factor=1.0, ’-shift’, shift=0.0, ’-zeroShift’, ze-
roShift=0.0)
This command is used to construct a TimeSeries object in which the load factor is some trigonemtric function
of the time in the domain
{︃
𝑐𝐹 𝑎𝑐𝑡𝑜𝑟 * 𝑠𝑖𝑛( 2.0𝜋(𝑡−𝑡𝑆𝑡𝑎𝑟𝑡)
𝑝𝑒𝑟𝑖𝑜𝑑 + 𝜑), 𝑡𝑆𝑡𝑎𝑟𝑡 <= 𝑡 <= 𝑡𝐸𝑛𝑑
𝜆 = 𝑓 (𝑡) =
0.0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
𝑝𝑒𝑟𝑖𝑜𝑑 𝑧𝑒𝑟𝑜𝑆ℎ𝑖𝑓 𝑡
𝜑 = 𝑠ℎ𝑖𝑓 𝑡 − * arcsin( )
2.0𝜋 𝑐𝐹 𝑎𝑐𝑡𝑜𝑟

104 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

tag (int) unique tag among TimeSeries objects.

tStart (float) Starting time of non-zero load factor.
tEnd (float) Ending time of non-zero load factor.
period (float) Characteristic period of sine wave.
shift (float) Phase shift in radians. (optional)
factor (float) Load factor. (optional)
zeroShift (float) Zero shift. (optional)

Triangular TimeSeries

timeSeries(’Triangle’, tag, tStart, tEnd, period, ’-factor’, factor=1.0, ’-shift’, shift=0.0, ’-zeroShift’, ze-
roShift=0.0)
This command is used to construct a TimeSeries object in which the load factor is some triangular function of
the time in the domain.
⎧
⎪
⎪ 𝑠𝑙𝑜𝑝𝑒 * 𝑘 * 𝑝𝑒𝑟𝑖𝑜𝑑 + 𝑧𝑒𝑟𝑜𝑆ℎ𝑖𝑓 𝑡, 𝑘 < 0.25
⎪
⎨𝑐𝐹 𝑎𝑐𝑡𝑜𝑟 − 𝑠𝑙𝑜𝑝𝑒 * (𝑘 − 0.25) * 𝑝𝑒𝑟𝑖𝑜𝑑 + 𝑧𝑒𝑟𝑜𝑆ℎ𝑖𝑓 𝑡, 𝑘 < 0.75
𝜆 = 𝑓 (𝑡) =
⎪
⎪
⎪ −𝑐𝐹 𝑎𝑐𝑡𝑜𝑟 + 𝑠𝑙𝑜𝑝𝑒 * (𝑘 − 0.75) * 𝑝𝑒𝑟𝑖𝑜𝑑 + 𝑧𝑒𝑟𝑜𝑆ℎ𝑖𝑓 𝑡, 𝑘 < 1.0
0.0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
⎩

𝑐𝐹 𝑎𝑐𝑡𝑜𝑟
𝑠𝑙𝑜𝑝𝑒 =
𝑝𝑒𝑟𝑖𝑜𝑑/4
𝑡 + 𝜑 − 𝑡𝑆𝑡𝑎𝑟𝑡 𝑡 + 𝜑 − 𝑡𝑆𝑡𝑎𝑟𝑡
𝑘= − 𝑓 𝑙𝑜𝑜𝑟( )
𝑝𝑒𝑟𝑖𝑜𝑑 𝑝𝑒𝑟𝑖𝑜𝑑
𝑧𝑒𝑟𝑜𝑆ℎ𝑖𝑓 𝑡
𝜑 = 𝑠ℎ𝑖𝑓 𝑡 −
𝑠𝑙𝑜𝑝𝑒

tag (int) unique tag among TimeSeries objects.

tStart (float) Starting time of non-zero load factor.
tEnd (float) Ending time of non-zero load factor.
period (float) Characteristic period of sine wave.
shift (float) Phase shift in radians. (optional)
factor (float) Load factor. (optional)
zeroShift (float) Zero shift. (optional)

Rectangular TimeSeries

timeSeries(’Rectangular’, tag, tStart, tEnd, ’-factor’, factor=1.0)

This command is used to construct a TimeSeries object in which the load factor is constant for a specified period
and 0 otherwise, i.e.
{︃
𝑐𝐹 𝑎𝑐𝑡𝑜𝑟, 𝑡𝑆𝑡𝑎𝑟𝑡 <= 𝑡 <= 𝑡𝐸𝑛𝑑
𝜆 = 𝑓 (𝑡) =
0.0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒

tag (int) unique tag among TimeSeries objects.

tStart (float) Starting time of non-zero load factor.
tEnd (float) Ending time of non-zero load factor.
factor (float) Load factor. (optional)

1.4. Model Commands 105

OpenSeesPy Documentation, Release 3.5.1.3

Pulse TimeSeries

timeSeries(’Pulse’, tag, tStart, tEnd, period, ’-width’, width=0.5, ’-shift’, shift=0.0, ’-factor’, factor=1.0,
’-zeroShift’, zeroShift=0.0)
This command is used to construct a TimeSeries object in which the load factor is some pulse function of the
time in the domain.
⎧
⎨𝑐𝐹 𝑎𝑐𝑡𝑜𝑟 + 𝑧𝑒𝑟𝑜𝑆ℎ𝑖𝑓 𝑡, 𝑘 < 𝑤𝑖𝑑𝑡ℎ
⎪
𝜆 = 𝑓 (𝑡) = 𝑧𝑒𝑟𝑜𝑠ℎ𝑖𝑓 𝑡, 𝑘<1
⎪
0.0, 𝑜𝑡ℎ𝑒𝑟𝑤𝑖𝑠𝑒
⎩

𝑡 + 𝑠ℎ𝑖𝑓 𝑡 − 𝑡𝑆𝑡𝑎𝑟𝑡 𝑡 + 𝑠ℎ𝑖𝑓 𝑡 − 𝑡𝑆𝑡𝑎𝑟𝑡

𝑘= − 𝑓 𝑙𝑜𝑜𝑟( )
𝑝𝑒𝑟𝑖𝑜𝑑 𝑝𝑒𝑟𝑖𝑜𝑑

tag (int) unique tag among TimeSeries objects.

tStart (float) Starting time of non-zero load factor.
tEnd (float) Ending time of non-zero load factor.
period (float) Characteristic period of pulse.
width (float) Pulse width as a fraction of the period. (optinal)
shift (float) Phase shift in seconds. (optional)
factor (float) Load factor. (optional)
zeroShift (float) Zero shift. (optional)

Path TimeSeries

timeSeries(’Path’, tag, ’-dt’, dt=0.0, ’-values’, *values, ’-time’, *time, ’-filePath’, filePath=”, ’-fileTime’,
fileTime=”, ’-factor’, factor=1.0, ’-startTime’, startTime=0.0, ’-useLast’, ’-prependZero’)
The relationship between load factor and time is input by the user as a series of discrete points in the 2d space
(load factor, time). The input points can come from a file or from a list in the script. When the time specified
does not match any of the input points, linear interpolation is used between points. There are many ways to
specify the load path, for example, the load factors set with values or filePath, and the time set with dt,
time, or fileTime.

tag (int) unique tag among TimeSeries objects.

dt (float) Time interval between specified points. (optional)
values (list (float)) Load factor values in a (list). (optional)
time (list (float)) Time values in a (list). (optional)
filePath (str) File containing the load factors values. (optional)
fileTime (str) File containing the time values for corresponding load factors. (optional)
factor (float) A factor to multiply load factors by. (optional)
startTime (float) Provide a start time for provided load factors. (optional)
'-useLast' (str) Use last value after the end of the series. (optional)
'-prependZero' (str) Prepend a zero value to the series of load factors. (optional)

• Linear interpolation between points.

• If the specified time is beyond last point (AND WATCH FOR NUMERICAL ROUNDOFF), 0.0 is re-
turned. Specify '-useLast' to use the last data point instead of 0.0.
• The transient integration methods in OpenSees assume zero initial conditions. So it is important that any
timeSeries that is being used in a transient analysis‘ starts from zero (first data point in the timeSeries =
0.0). To guarantee that this is the case the optional parameter '-prependZero' can be specified to
prepend a zero value to the provided TimeSeries.

106 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1.4.8 pattern commands

pattern(patternType, patternTag, *patternArgs)

The pattern command is used to construct a LoadPattern and add it to the Domain. Each LoadPattern in OpenSees has
a TimeSeries associated with it. In addition it may contain ElementLoads, NodalLoads and SinglePointConstraints.
Some of these SinglePoint constraints may be associated with GroundMotions.

patternType (str) pattern type.

patternTag (int) pattern tag.
patternArgs (list) a list of pattern arguments

The following contain information about available patternType:

1. Plain Pattern
2. UniformExcitation Pattern
3. Multi-Support Excitation Pattern

Plain Pattern

pattern(’Plain’, patternTag, tsTag, ’-fact’, fact)

This commnand allows the user to construct a LoadPattern object. Each plain load pattern is associated with
a TimeSeries object and can contain multiple NodalLoads, ElementalLoads and SP_Constraint objects. The
command to generate LoadPattern object contains in { } the commands to generate all the loads and the single-
point constraints in the pattern. To construct a load pattern and populate it, the following command is used:

patternTag (int) unique tag among load patterns.

tsTag (int) the tag of the time series to be used in the load pattern
fact (float) constant factor. (optional)

Note: the commands below to generate all the loads and sp constraints will be included in last called pattern command.

load command

load(nodeTag, *loadValues)
This command is used to construct a NodalLoad object and add it to the enclosing LoadPattern.

nodeTag (int) tag of node to which load is applied.

loadValues (list (float)) ndf reference load values.

Note: The load values are reference loads values. It is the time series that provides the load factor. The load factor
times the reference values is the load that is actually applied to the node.

1.4. Model Commands 107

OpenSeesPy Documentation, Release 3.5.1.3

eleLoad command

eleLoad(’-ele’, *eleTags, ’-range’, eleTag1, eleTag2, ’-type’, ’-beamUniform’, Wy, <Wz>, Wx=0.0, ’-
beamPoint’, Py, <Pz>, xL, Px=0.0, ’-beamThermal’, *tempPts)
The eleLoad command is used to construct an ElementalLoad object and add it to the enclosing LoadPattern.

eleTags tag of PREVIOUSLY DEFINED element

(list (int))
eleTag1 element tag
(int)
eleTag2 element tag
(int)
Wx (float) mag of uniformily distributed ref load acting in direction along member length. (optional)
Wy (float) mag of uniformily distributed ref load acting in local y direction of element
Wz (float) mag of uniformily distributed ref load acting in local z direction of element. (required only for
3D)
Px (float) mag of ref point load acting in direction along member length. (optional)
Py (float) mag of ref point load acting in local y direction of element
Pz (float) mag of ref point load acting in local z direction of element. (required only for 3D)
xL (float) location of point load relative to node I, prescribed as fraction of element length
tempPts temperature points: temPts = [T1, y1, T2, y2, ..., T9, y9] Each point
(list (T1, y1) define a temperature and location. This command may accept 2,5 or 9 temper-
(float)) ature points.

Note:
1. The load values are reference load values, it is the time series that provides the load factor. The load factor times
the reference values is the load that is actually applied to the element.
2. At the moment, eleLoads do not work with 3D beam-column elements if Corotational geometric transformation
is used.

sp command

sp(nodeTag, dof, dofValue)

This command is used to construct a single-point constraint object and add it to the enclosing LoadPattern.

nodeTag (int) tag of node to which load is applied.

dof (int) the degree-of-freedom at the node to which constraint is applied (1 through ndf)
dofValue (float) reference constraint value.

Note: The dofValue is a reference value, it is the time series that provides the load factor. The load factor times the
reference value is the constraint that is actually applied to the node.

108 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

UniformExcitation Pattern

pattern(’UniformExcitation’, patternTag, dir, ’-disp’, dispSeriesTag, ’-vel’, velSeriesTag, ’-accel’, ac-

celSeriesTag, ’-vel0’, vel0, ’-fact’, fact)
The UniformExcitation pattern allows the user to apply a uniform excitation to a model acting in a certain
direction. The command is as follows:

patternTag (int) unique tag among load patterns

dir (int) direction in which ground motion acts
1. corresponds to translation along the global X
axis
2. corresponds to translation along the global Y
axis
3. corresponds to translation along the global Z
axis
4. corresponds to rotation about the global X axis
5. corresponds to rotation about the global Y axis
6. corresponds to rotation about the global Z axis

dispSeriesTag (int) tag of the TimeSeries series defining the displace-

ment history. (optional)
velSeriesTag (int) tag of the TimeSeries series defining the velocity his-
tory. (optional)
accelSeriesTag (int) tag of the TimeSeries series defining the acceleration
history. (optional)
vel0 (float) the initial velocity (optional, default=0.0)
fact (float) constant factor (optional, default=1.0)

Note:
1. The responses obtained from the nodes for this type of excitation are RELATIVE values, and not the absolute
values obtained from a multi-support case.
2. must set one of the disp, vel or accel time series

Multi-Support Excitation Pattern

pattern(’MultipleSupport’, patternTag)
The Multi-Support pattern allows similar or different prescribed ground motions to be input at various supports
in the structure. In OpenSees, the prescribed motion is applied using single-point constraints, the single-point
constraints taking their constraint value from user created ground motions.

patternTag (int) integer tag identifying pattern

Note:
1. The results for the responses at the nodes are the ABSOLUTE values, and not relative values as in the case of a
UniformExciatation.
2. The non-homogeneous single point constraints require an appropriate choice of constraint handler.

1.4. Model Commands 109

OpenSeesPy Documentation, Release 3.5.1.3

Plain Ground Motion

groundMotion(gmTag, ’Plain’, ’-disp’, dispSeriesTag, ’-vel’, velSeriesTag, ’-accel’, accelSeriesTag, ’-int’,

tsInt=’Trapezoidal’, ’-fact’, factor=1.0)
This command is used to construct a plain GroundMotion object. Each GroundMotion object is associated
with a number of TimeSeries objects, which define the acceleration, velocity and displacement records for that
ground motion. T

gmTag (int) unique tag among ground motions in load pattern

dispSeriesTag (int) tag of the TimeSeries series defining the displacement history. (optional)
velSeriesTag (int) tag of the TimeSeries series defining the velocity history. (optional)
accelSeriesTag (int) tag of the TimeSeries series defining the acceleration history. (optional)
tsInt (str) 'Trapezoidal' or 'Simpson' numerical integration method
factor (float) constant factor. (optional)

Note:
1. The displacements are the ones used in the ImposedMotions to set nodal response.
2. If only the acceleration TimeSeries is provided, numerical integration will be used to determine the velocities
and displacements.
3. For earthquake excitations it is important that the user provide the displacement time history, as the one gener-
ated using the trapezoidal method will not provide good results.
4. Any combination of the acceleration, velocity and displacement time-series can be specified.

Interpolated Ground Motion

groundMotion(gmTag, ’Interpolated’, *gmTags, ’-fact’, facts)

This command is used to construct an interpolated GroundMotion object, where the motion is determined by
combining several previously defined ground motions in the load pattern.

gmTag (int) unique tag among ground motions in load pattern

gmTags (list (int)) the tags of existing ground motions in pattern to be used for interpolation
facts (list (float)) the interpolation factors. (optional)

Imposed Motion

imposedMotion(nodeTag, dof, gmTag)

This command is used to construct an ImposedMotionSP constraint which is used to enforce the response
of a dof at a node in the model. The response enforced at the node at any give time is obtained from the
GroundMotion object associated with the constraint.

nodeTag (int) tag of node on which constraint is to be placed

dof (int) dof of enforced response. Valid range is from 1 through ndf at node.
gmTag (int) pre-defined GroundMotion object tag

110 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1.4.9 mass command

mass(nodeTag, *massValues)
This command is used to set the mass at a node, replacing any previously defined mass at the node.

nodeTag (int) integer tag identifying node whose mass is set

massValues (list (float)) ndf nodal mass values corresponding to each DOF

1.4.10 region command

region(regTag, ’-ele’, *eles, ’-eleOnly’, *eles, ’-eleRange’, startEle, endEle, ’-eleOnlyRange’, startEle, en-
dEle, ’-node’, *nodes, ’-nodeOnly’, *nodes, ’-nodeRange’, startNode, endNode, ’-nodeOnlyRange’,
startNode, endNode, ’-rayleigh’, alphaM, betaK, betaKinit, betaKcomm)
The region command is used to label a group of nodes and elements. This command is also used to assign
rayleigh damping parameters to the nodes and elements in this region. The region is specified by either elements
or nodes, not both. If elements are defined, the region includes these elements and the all connected nodes, unless
the -eleOnly option is used in which case only elements are included. If nodes are specified, the region includes
these nodes and all elements of which all nodes are prescribed to be in the region, unless the -nodeOnly option
is used in which case only the nodes are included.

regTag (int) unique integer tag

eles (list (int)) tags of selected elements in domain to be included in region (optional)
nodes (list (int)) tags of selected nodes in domain to be included in region (optional)
startEle (int) tag for start element (optional)
endEle (int) tag for end element (optional)
startNode (int) tag for start node (optional)
endNode (int) tag for end node (optional)
alphaM (float) factor applied to elements or nodes mass matrix (optional)
betaK (float) factor applied to elements current stiffness matrix (optional)
betaKinit (float) factor applied to elements initial stiffness matrix (optional)
betaKcomm (float) factor applied to elements committed stiffness matrix (optional)

Note: The user cannot prescribe the region by BOTH elements and nodes.

1.4.11 rayleigh command

rayleigh(alphaM, betaK, betaKinit, betaKcomm)

This command is used to assign damping to all previously-defined elements and nodes. When using rayleigh
damping in OpenSees, the damping matrix for an element or node, D is specified as a combination of stiffness
and mass-proportional damping matrices:

𝐷 = 𝛼𝑀 * 𝑀 + 𝛽𝐾 * 𝐾𝑐𝑢𝑟𝑟 + 𝛽𝐾𝑖𝑛𝑖𝑡 * 𝐾𝑖𝑛𝑖𝑡 + 𝛽𝐾𝑐𝑜𝑚𝑚 * 𝐾𝑐𝑜𝑚𝑚𝑖𝑡

alphaM (float) factor applied to elements or nodes mass matrix

betaK (float) factor applied to elements current stiffness matrix.
betaKinit (float) factor applied to elements initial stiffness matrix.
betaKcomm (float) factor applied to elements committed stiffness matrix.

1.4. Model Commands 111

OpenSeesPy Documentation, Release 3.5.1.3

1.4.12 block commands

Create a block of mesh

block2D command

block2D(numX, numY, startNode, startEle, eleType, *eleArgs, *crds)

Create mesh of quadrilateral elements

numX number of elements in local x directions of the block.

(int)
numY number of elements in local y directions of the block.
(int)
node from which the mesh generation will start.
startNode
(int)
element from which the mesh generation will start.
startEle
(int)
eleTypeelement type ('quad', 'shell', 'bbarQuad', 'enhancedQuad', or 'SSPquad')
(str)
eleArgsa list of element parameters.
(list)
crds coordinates of the block elements with the format:
(list) [1, x1, y1, <z1>,
2, x2, y2, <z2>,
3, x3, y3, <z3>,
4, x4, y4, <z4>,
<5>, <x5>, <y5>, <z5>,
<6>, <x6>, <y6>, <z6>,
<7>, <x7>, <y7>, <z7>,
<8>, <x8>, <y8>, <z8>,
<9>, <x9>, <y9>, <z9>]
<> means optional

block3D command

block3D(numX, numY, numZ, startNode, startEle, eleType, *eleArgs, *crds)

Create mesh of quadrilateral elements

112 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

numX number of elements in local x directions of the block.

(int)
numY number of elements in local y directions of the block.
(int)
numZ number of elements in local z directions of the block.
(int)
node from which the mesh generation will start.
startNode
(int)
element from which the mesh generation will start.
startEle
(int)
eleTypeelement type ('stdBrick', 'bbarBrick', 'Brick20N')
(str)
eleArgsa list of element parameters.
(list)
crds coordinates of the block elements with the format:
(list) [1, x1, y1, z1,
2, x2, y2, z2,
3, x3, y3, z3,
4, x4, y4, z4,
5, x5, y5, z5,
6, x6, y6, z6,
7, x7, y7, z7,
8, x8, y8, z8,
9, x9, y9, z9,
<10>, <x10>, <y10>, <z10>,
<11>, <x11>, <y11>, <z11>,
<12>, <x12>, <y12>, <z12>,
<13>, <x13>, <y13>, <z13>,
<14>, <x14>, <y14>, <z14>,
<15>, <x15>, <y15>, <z15>,
<16>, <x16>, <y16>, <z16>,
<17>, <x17>, <y17>, <z17>,
<18>, <x18>, <y18>, <z18>,
<19>, <x19>, <y19>, <z19>,
<20>, <x20>, <y20>, <z20>,
<21>, <x21>, <y21>, <z21>,
<22>, <x22>, <y22>, <z22>,
<23>, <x23>, <y23>, <z23>,
<24>, <x24>, <y24>, <z24>,
<25>, <x25>, <y25>, <z25>,
<26>, <x26>, <y26>, <z26>,
<27>, <x27>, <y27>, <z27>]
<> means optional

1.4.13 beamIntegration commands

beamIntegration(type, tag, *args)

A wide range of numerical integration options are available in OpenSees to represent distributed plasticity or
non-prismatic section details in Beam-Column Elements, i.e., across the entire element domain [0, L].
Following are beamIntegration types available in the OpenSees:
Integration Methods for Distributed Plasticity. Distributed plasticity methods permit yielding at any integration point

1.4. Model Commands 113

OpenSeesPy Documentation, Release 3.5.1.3

along the element length.

1. Lobatto
2. Legendre
3. NewtonCotes
4. Radau
5. Trapezoidal
6. CompositeSimpson
7. UserDefined
8. FixedLocation
9. LowOrder
10. MidDistance

Lobatto

beamIntegration(’Lobatto’, tag, secTag, N)

Create a Gauss-Lobatto beamIntegration object. Gauss-Lobatto integration is the most common approach for
evaluating the response of forceBeamColumn (Neuenhofer and Filippou 1997) because it places an integration
point at each end of the element, where bending moments are largest in the absence of interior element loads.

tag (int) tag of the beam integration.

secTag (int) A previous-defined section object.
N (int) Number of integration points along the element.

Legendre

beamIntegration(’Legendre’, tag, secTag, N)

Create a Gauss-Legendre beamIntegration object. Gauss-Legendre integration is more accurate than Gauss-
Lobatto; however, it is not common in force-based elements because there are no integration points at the
element ends.
Places N Gauss-Legendre integration points along the element. The location and weight of each integration point
are tabulated in references on numerical analysis. The force deformation response at each integration point is
defined by the section. The order of accuracy for Gauss-Legendre integration is 2N-1.
Arguments and examples see Lobatto.

NewtonCotes

beamIntegration(’NewtonCotes’, tag, secTag, N)

Create a Newton-Cotes beamIntegration object. Newton-Cotes places integration points uniformly along the
element, including a point at each end of the element.
Places N Newton-Cotes integration points along the element. The weights for the uniformly spaced integration
points are tabulated in references on numerical analysis. The force deformation response at each integration
point is defined by the section. The order of accuracy for Gauss-Radau integration is N-1.
Arguments and examples see Lobatto.

114 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Radau

beamIntegration(’Radau’, tag, secTag, N)

Create a Gauss-Radau beamIntegration object. Gauss-Radau integration is not common in force-based elements
because it places an integration point at only one end of the element; however, it forms the basis for optimal
plastic hinge integration methods.
Places N Gauss-Radau integration points along the element with a point constrained to be at ndI. The location
and weight of each integration point are tabulated in references on numerical analysis. The force-deformation
response at each integration point is defined by the section. The order of accuracy for Gauss-Radau integration
is 2N-2.
Arguments and examples see Lobatto.

Trapezoidal

beamIntegration(’Trapezoidal’, tag, secTag, N)

Create a Trapezoidal beamIntegration object.
Arguments and examples see Lobatto.

CompositeSimpson

beamIntegration(’CompositeSimpson’, tag, secTag, N)

Create a CompositeSimpson beamIntegration object.
Arguments and examples see Lobatto.

UserDefined

beamIntegration(’UserDefined’, tag, N, *secTags, *locs, *wts)

Create a UserDefined beamIntegration object. This option allows user-specified locations and weights of the
integration points.

tag (int) tag of the beam integration

N (int) number of integration points along the element.
secTags (list (int)) A list previous-defined section objects.
locs (list (float)) Locations of integration points along the element.
wts (list (float)) weights of integration points.

locs = [0.1, 0.3, 0.5, 0.7, 0.9]

wts = [0.2, 0.15, 0.3, 0.15, 0.2]
secs = [1, 2, 2, 2, 1]
beamIntegration('UserDefined',1,len(secs),*secs,*locs,*wts)

Places N integration points along the element, which are defined in locs on the natural domain [0, 1]. The
weight of each integration point is defined in the wts also on the [0, 1] domain. The force-deformation response
at each integration point is defined by the secs. The locs, wts, and secs should be of length N. In general,
there is no accuracy for this approach to numerical integration.

1.4. Model Commands 115

OpenSeesPy Documentation, Release 3.5.1.3

FixedLocation

beamIntegration(’FixedLocation’, tag, N, *secTags, *locs)

Create a FixedLocation beamIntegration object. This option allows user-specified locations of the integration
points. The associated integration weights are computed by the method of undetermined coefficients (Vander-
monde system)
𝑁 ∫︁ 1
∑︁ 1
𝑥𝑗−1
𝑖 𝑤𝑖 = 𝑥𝑗−1 𝑑𝑥 = , (𝑗 = 1, ..., 𝑁 )
𝑖=1 0 𝑗

Note that NewtonCotes integration is recovered when the integration point locations are equally spaced.

tag (int) tag of the beam integration

N (int) number of integration points along the element.
secTags (list (int)) A list previous-defined section objects.
locs (list (float)) Locations of integration points along the element.

Places N integration points along the element, whose locations are defined in locs. on the natural domain [0,
1]. The force-deformation response at each integration point is defined by the secs. Both the locs and secs
should be of length N. The order of accuracy for Fixed Location integration is N-1.

LowOrder

beamIntegration(’LowOrder’, tag, N, *secTags, *locs, *wts)

Create a LowOrder beamIntegration object. This option is a generalization of the FixedLocation and UserDe-
fined integration approaches and is useful for moving load analysis (Kidarsa, Scott and Higgins 2008). The
locations of the integration points are user defined, while a selected number of weights are specified and the
remaining weights are computed by the method of undetermined coefficients.
𝑁𝑓 𝑐 𝑁
∑︁ 1 ∑︁
𝑥𝑗−1
𝑓 𝑖 𝑤𝑓 𝑖 = − 𝑥𝑗−1 𝑤𝑐𝑖
𝑖=1
𝑗 𝑖=1 𝑐𝑖

Note that FixedLocation integration is recovered when Nc is zero.

tag (int) tag of the beam integration

N (int) number of integration points along the element.
secTags (list (int)) A list previous-defined section objects.
locs (list (float)) Locations of integration points along the element.
wts (list (float)) weights of integration points.

locs = [0.0, 0.2, 0.5, 0.8, 1.0]

wts = [0.2, 0.2]
secs = [1, 2, 2, 2, 1]
beamIntegration('LowOrder',1,len(secs),*secs,*locs,*wts)

Places N integration points along the element, which are defined in locs. on the natural domain [0, 1]. The
force-deformation response at each integration point is defined by the secs. Both the locs and secs should
be of length N. The wts at user-selected integration points are specified on [0, 1], which can be of length Nc
equals 0 up to N. These specified weights are assigned to the first Nc entries in the locs and secs, respectively.
The order of accuracy for Low Order integration is N-Nc-1.

116 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Note: Nc is determined from the length of the wts list. Accordingly, FixedLocation integration is recovered
when wts is an empty list and UserDefined integration is recovered when the wts and locs lists are of equal
length.

MidDistance

beamIntegration(’MidDistance’, tag, N, *secTags, *locs)

Create a MidDistance beamIntegration object. This option allows user-specified locations of the integration
points. The associated integration weights are determined from the midpoints between adjacent integration
point locations. 𝑤𝑖 = (𝑥𝑖+1 − 𝑥𝑖−1 )/2 for 𝑖 = 2...𝑁 − 1, 𝑤1 = (𝑥1 + 𝑥2 )/2, and 𝑤𝑁 = 1 − (𝑥𝑁 −1 + 𝑥𝑁 )/2.

tag (int) tag of the beam integration

N (int) number of integration points along the element.
secTags (list (int)) A list previous-defined section objects.
locs (list (float)) Locations of integration points along the element.

locs = [0.0, 0.2, 0.5, 0.8, 1.0]

secs = [1,2,2,2,1]
beamIntegration('MidDistance',1,len(secs),*secs,*locs)

Places N integration points along the element, whose locations are defined in locs on the natural domain [0,
1]. The force-deformation response at each integration point is defined by the secs. Both the locs and secs
should be of length N. This integration rule can only integrate constant functions exactly since the sum of the
integration weights is one.
For the locs shown above, the associated integration weights will be [0.15, 0.2, 0.3, 0.2, 0.15].
Plastic Hinge Integration Methods. Plastic hinge integration methods confine material yielding to regions of the
element of specified length while the remainder of the element is linear elastic. A summary of plastic hinge integration
methods is found in (Scott and Fenves 2006).
1. UserHinge
2. HingeMidpoint
3. HingeRadau
4. HingeRadauTwo
5. HingeEndpoint

UserHinge

beamIntegration(’UserHinge’, tag, secETag, npL, *secsLTags, *locsL, *wtsL, npR, *secsRTags, *locsR,
*wtsR)
Create a UserHinge beamIntegration object.

1.4. Model Commands 117

OpenSeesPy Documentation, Release 3.5.1.3

tag (int) tag of the beam integration

secE (int) A previous-defined section tags for element interior
npI (int) number of integration points along the hinge at end I
secsI (list (int)) A list of previous-defined section tags for hinge at end I
locsI (list (float)) A list of locations of integration points for hinge at end I
wtsI (list (float)) A list of weights of integration points for hinge at end I
npJ (int) number of integration points along the hinge at end J
secsJ (list (int)) A list of previous-defined section tags for hinge at end J
locsJ (list (float)) A list of locations of integration points for hinge at end J
wtsJ (list (float)) A list of weights of integration points for hinge at end J

tag = 1
secE = 5

npI = 2
secsI = [1,2]
locsI = [0.1,0.2]
wtsI = [0.1,0.05]

npJ = 2
secsJ = [3,4]
locsJ = [0.8,0.9]
wtsJ = [0.05,0.1]

beamIntegration('UserHinge',tag,secE,npI,*secsI,*locsI,*wtsI,npJ,*secsJ,*locsJ,
˓→*wtsJ)

HingeMidpoint

beamIntegration(’HingeMidpoint’, tag, secI, lpI, secJ, lpJ, secE)

Create a HingeMidpoint beamIntegration object. Midpoint integration over each hinge region is the most accu-
rate one-point integration rule; however, it does not place integration points at the element ends and there is a
small integration error for linear curvature distributions along the element.

tag (int) tag of the beam integration.

secI (int) A previous-defined section object for hinge at I.
lpI (float) The plastic hinge length at I.
secJ (int) A previous-defined section object for hinge at J.
lpJ (float) The plastic hinge length at J.
secE (int) A previous-defined section object for the element interior.

The plastic hinge length at end I (J) is equal to lpI (lpJ) and the associated force deformation response is
defined by the secI (secJ). The force deformation response of the element interior is defined by the secE.
Typically, the interior section is linear-elastic, but this is not necessary.

lpI = 0.1
lpJ = 0.2
beamIntegration('HingeMidpoint',tag,secI,lpI,secJ,lpJ,secE)

118 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

HingeRadau

beamIntegration(’HingeRadau’, tag, secI, lpI, secJ, lpJ, secE)

Create a HingeRadau beamIntegration object. Modified two-point Gauss-Radau integration over each hinge re-
gion places an integration point at the element ends and at 8/3 the hinge length inside the element. This approach
represents linear curvature distributions exactly and the characteristic length for softening plastic hinges is equal
to the assumed palstic hinge length.
Arguments and examples see HingeMidpoint.

HingeRadauTwo

beamIntegration(’HingeRadauTwo’, tag, secI, lpI, secJ, lpJ, secE)

Create a HingeRadauTwo beamIntegration object. Two-point Gauss-Radau integration over each hinge region
places an integration point at the element ends and at 2/3 the hinge length inside the element. This approach
represents linear curvature distributions exactly; however, the characteristic length for softening plastic hinges
is not equal to the assumed plastic hinge length (equals 1/4 of the plastic hinge length).
Arguments and examples see HingeMidpoint.

HingeEndpoint

beamIntegration(’HingeEndpoint’, tag, secI, lpI, secJ, lpJ, secE)

Create a HingeEndpoint beamIntegration object. Endpoint integration over each hinge region moves the inte-
gration points to the element ends; however, there is a large integration error for linear curvature distributions
along the element.

tag (int) tag of the beam integration.

secI (int) A previous-defined section object for hinge at I.
lpI (float) The plastic hinge length at I.
secJ (int) A previous-defined section object for hinge at J.
lpJ (float) The plastic hinge length at J.
secE (int) A previous-defined section object for the element interior.

Arguments and examples see HingeMidpoint.

1.4.14 uniaxialMaterial commands

uniaxialMaterial(matType, matTag, *matArgs)

This command is used to construct a UniaxialMaterial object which represents uniaxial stress-strain (or force-
deformation) relationships.

matType (str) material type

matTag (int) material tag.
matArgs (list) a list of material arguments, must be preceded with *.

For example,
matType = 'Steel01'
matTag = 1
matArgs = [Fy, E0, b]
uniaxialMaterial(matType, matTag, *matArgs)

1.4. Model Commands 119

OpenSeesPy Documentation, Release 3.5.1.3

The following contain information about available matType:

Steel & Reinforcing-Steel Materials

1. Steel01
2. Steel02
3. Steel4
4. Hysteretic
5. ReinforcingSteel
6. Dodd_Restrepo
7. RambergOsgoodSteel
8. SteelMPF
9. Steel01Thermal

Steel01

uniaxialMaterial(’Steel01’, matTag, Fy, E0, b, a1, a2, a3, a4)

This command is used to construct a uniaxial bilinear steel material object with kinematic hardening and optional
isotropic hardening described by a non-linear evolution equation (REF: Fedeas).

matTag integer tag identifying material

(int)
Fy yield strength
(float)
E0 initial elastic tangent
(float)
b (float) strain-hardening ratio (ratio between post-yield tangent and initial elastic tangent)
a1 isotropic hardening parameter, increase of compression yield envelope as proportion of yield
(float) strength after a plastic strain of 𝑎2 * (𝐹𝑦 /𝐸0 ) (optional)
a2 isotropic hardening parameter (see explanation under a1). (optional).
(float)
a3 isotropic hardening parameter, increase of tension yield envelope as proportion of yield strength
(float) after a plastic strain of 𝑎4 * (𝐹𝑦 /𝐸0 ). (optional)
a4 isotropic hardening parameter (see explanation under a3). (optional)
(float)

Note: If strain-hardening ratio is zero and you do not expect softening of your system use BandSPD solver.

Steel02

uniaxialMaterial(’Steel02’, matTag, Fy, E0, b, *params, a1=a2*Fy/E0, a2=1.0, a3=a4*Fy/E0, a4=1.0,

sigInit=0.0)
This command is used to construct a uniaxial Giuffre-Menegotto-Pinto steel material object with isotropic strain
hardening.

120 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

Fy (float) yield strength
E0 (float) initial elastic tangent
b (float) strain-hardening ratio (ratio between post-yield tan-
gent and initial elastic tangent)
params (list (float)) parameters to control the transition from elastic
to plastic branches. params=[R0,cR1,cR2].
Recommended values: R0=between 10 and 20,
cR1=0.925, cR2=0.15
a1 (float) isotropic hardening parameter, increase of compres-
sion yield envelope as proportion of yield strength
after a plastic strain of 𝑎2 * (𝐹𝑦 /𝐸0 ) (optional)
a2 (float) isotropic hardening parameter (see explanation un-
der a1). (optional).
a3 (float) isotropic hardening parameter, increase of tension
yield envelope as proportion of yield strength after
a plastic strain of 𝑎4 * (𝐹𝑦 /𝐸0 ). (optional)
a4 (float) isotropic hardening parameter (see explanation un-
der a3). (optional)
sigInit (float) Initial Stress Value (optional, default: 0.0) the strain
is calculated from epsP=sigInit/E
if (sigInit!= 0.0) {
double epsInit = sigInit/E;
eps = trialStrain+epsInit;
} else {
eps = trialStrain;
}

See also:
Steel02

Steel4

uniaxialMaterial(’Steel4’, matTag, Fy, E0, ’-asym’, ’-kin’, b_k, *params, b_kc, R_0c, r_1c, r_2c, ’-iso’,
b_i, rho_i, b_l, R_i, l_yp, b_ic, rho_ic, b_lc, R_ic, ’-ult’, f_u, R_u, f_uc, R_uc, ’-init’,
sig_init, ’-mem’, cycNum)
This command is used to construct a general uniaxial material with combined kinematic and isotropic hardening
and optional non-symmetric behavior.

1.4. Model Commands 121

OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
Fy yield strength
(float)
E0 initial elastic tangent
(float)
'-kin' apply kinematic hardening
(str)
b_k hardening ratio (E_k/E_0)
(float)
params control the exponential transition from linear elastic to hardening asymptote params=[R_0,
(list r_1,r_2]. Recommended values: R_0 = 20, r_1 = 0.90, r_2 = 0.15
(float))
'-iso' apply isotropic hardening
(str)
b_i initial hardening ratio (E_i/E_0)
(float)
b_l saturated hardening ratio (E_is/E_0)
(float)
rho_i specifies the position of the intersection point between initial and saturated hardening asymptotes
(float)
R_i control the exponential transition from initial to saturated asymptote
(float)
l_yp length of the yield plateau in eps_y0 = f_y / E_0 units
(float)
'-ult' apply an ultimate strength limit
(str)
f_u ultimate strength
(float)
R_u control the exponential transition from kinematic hardening to perfectly plastic asymptote
(float)
'-asym' assume non-symmetric behavior
(str)
'-init' apply initial stress
(str)
sig_initinitial stress value
(float)
'-mem' configure the load history memory
(str)
cycNum expected number of half-cycles during the loading process Efficiency of the material can be
(float) slightly increased by correctly setting this value. The default value is cycNum = 50 Load
history memory can be turned off by setting cycNum = 0.

See also:
Steel4

Hysteretic

uniaxialMaterial(’Hysteretic’, matTag, *p1, *p2, *p3=p2, *n1, *n2, *n3=n2, pinchX, pinchY, damage1,
damage2, beta=0.0)
This command is used to construct a uniaxial bilinear hysteretic material object with pinching of force and

122 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

deformation, damage due to ductility and energy, and degraded unloading stiffness based on ductility.

matTag integer tag identifying material

(int)
p1 (list p1=[s1p, e1p], stress and strain (or force & deformation) at first point of the envelope
(float)) in the positive direction
p2 (list p2=[s2p, e2p], stress and strain (or force & deformation) at second point of the enve-
(float)) lope in the positive direction
p3 (list p3=[s3p, e3p], stress and strain (or force & deformation) at third point of the envelope
(float)) in the positive direction
n1 (list n1=[s1n, e1n], stress and strain (or force & deformation) at first point of the envelope
(float)) in the negative direction
n2 (list n2=[s2n, e2n], stress and strain (or force & deformation) at second point of the enve-
(float)) lope in the negative direction
n3 (list n3=[s3n, e3n], stress and strain (or force & deformation) at third point of the envelope
(float)) in the negative direction
pinchX pinching factor for strain (or deformation) during reloading
(float)
pinchY pinching factor for stress (or force) during reloading
(float)
damage1 damage due to ductility: D1(mu-1)
(float)
damage2 damage due to energy: D2(Eii/Eult)
(float)
beta (float) power used to determine the degraded unloading stiffness based on ductility, mu-beta (op-
tional, default=0.0)

See also:
Hysteretic

ReinforcingSteel

uniaxialMaterial(’ReinforcingSteel’, matTag, fy, fu, Es, Esh, eps_sh, eps_ult, ’-GABuck’, lsr, beta, r,
gamma, ’-DMBuck’, lsr, alpha=1.0, ’-CMFatigue’, Cf, alpha, Cd, ’-IsoHard’, a1=4.3,
limit=1.0, ’-MPCurveParams’, R1=0.333, R2=18.0, R3=4.0)
This command is used to construct a ReinforcingSteel uniaxial material object. This object is intended to be
used in a reinforced concrete fiber section as the steel reinforcing material.

1.4. Model Commands 123

OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
fy (float) Yield stress in tension
fu (float) Ultimate stress in tension
Es (float) Initial elastic tangent
Esh (float) Tangent at initial strain hardening
eps_sh Strain corresponding to initial strain hardening
(float)
eps_ult Strain at peak stress
(float)
'-GABuck' Buckling Model Based on Gomes and Appleton (1997)
(str)
lsr (float) Slenderness Ratio
beta (float) Amplification factor for the buckled stress strain curve.
r (float) Buckling reduction factor
r can be a real number between [0.0 and 1.0]
r=1.0 full reduction (no buckling)
r=0.0 no reduction
0.0<r<1.0 linear interpolation between buckled and unbuckled curves
gamma Buckling constant
(float)
'-DMBuck' Buckling model based on Dhakal and Maekawa (2002)
(str)
lsr (float) Slenderness Ratio
alpha Adjustment Constant usually between 0.75 and 1.0 Default: alpha=1.0, this parameter is
(float) optional.
'-CMFatigue'Coffin-Manson Fatigue and Strength Reduction
(str)
Cf (float) Coffin-Manson constant C
alpha Coffin-Manson constant a
(float)
Cd (float) Cyclic strength reduction constant
'-IsoHard' Isotropic Hardening / Diminishing Yield Plateau
(str)
a1 (float) Hardening constant (default = 4.3)
limit Limit for the reduction of the yield plateau. % of original plateau length to remain (0.01 <
(float) limit < 1.0 ) Limit =1.0, then no reduction takes place (default =0.01)
Menegotto and Pinto Curve Parameters
'-MPCurveParams'
(str)
R1 (float) (default = 0.333)
R2 (float) (default = 18)
R3 (float) (default = 4)

See also:
Notes

Dodd_Restrepo

uniaxialMaterial(’Dodd_Restrepo’, matTag, Fy, Fsu, ESH, ESU, Youngs, ESHI, FSHI, OmegaFac=1.0)
This command is used to construct a Dodd-Restrepo steel material

124 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
Fy (float) Yield strength
Fsu Ultimate tensile strength (UTS)
(float)
ESH Tensile strain at initiation of strain hardening
(float)
ESU Tensile strain at the UTS
(float)
Youngs Modulus of elasticity
(float)
ESHI Tensile strain for a point on strain hardening curve, recommended range of values for ESHI: [
(float) (ESU + 5*ESH)/6, (ESU + 3*ESH)/4]
FSHI Tensile stress at point on strain hardening curve corresponding to ESHI
(float)
OmegaFac Roundedness factor for Bauschinger curve in cycle reversals from the strain hardening curve.
(float) Range: [0.75, 1.15]. Largest value tends to near a bilinear Bauschinger curve. Default = 1.0.

See also:
Notes

RambergOsgoodSteel

uniaxialMaterial(’RambergOsgoodSteel’, matTag, fy, E0, a, n)

This command is used to construct a Ramberg-Osgood steel material object.

matTag integer tag identifying material

(int)
fy Yield strength
(float)
E0 initial elastic tangent
(float)
a “yield offset” and the Commonly used value for a is 0.002
(float)
n Parameters to control the transition from elastic to plastic branches. And controls the hardening of
(float) the material by increasing the “n” hardening ratio will be decreased. Commonly used values for n
are ~5 or greater.

See also:
Notes

SteelMPF

uniaxialMaterial(’SteelMPF’, matTag, fyp, fyn, E0, bp, bn, *params, a1=0.0, a2=1.0, a3=0.0, a4=1.0)
This command is used to construct a uniaxialMaterial SteelMPF (Kolozvari et al., 2015), which represents the
well-known uniaxial constitutive nonlinear hysteretic material model for steel proposed by Menegotto and Pinto
(1973), and extended by Filippou et al. (1983) to include isotropic strain hardening effects.

1.4. Model Commands 125

OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
fyp Yield strength in tension (positive loading direction)
(float)
fyn Yield strength in compression (negative loading direction)
(float)
E0 (float) Initial tangent modulus
bp (float) Strain hardening ratio in tension (positive loading direction)
bn (float) Strain hardening ratio in compression (negative loading direction)
params parameters to control the transition from elastic to plastic branches. params=[R0,cR1,
(list cR2]. Recommended values: R0=20, cR1=0.925, cR2=0.15 or cR2=0.0015
(float))
a1 (float) Isotropic hardening in compression parameter (optional, default = 0.0). Shifts compression
yield envelope by a proportion of compressive yield strength after a maximum plastic tensile
strain of a2(fyp/E0)
a2 (float) Isotropic hardening in compression parameter (optional, default = 1.0).
a3 (float) Isotropic hardening in tension parameter (optional, default = 0.0). Shifts tension yield enve-
lope by a proportion of tensile yield strength after a maximum plastic compressive strain of
a3(fyn/E0).
a4 (float) Isotropic hardening in tension parameter (optional, default = 1.0). See explanation of a3.

See also:
Notes

Steel01Thermal

uniaxialMaterial(’Steel01Thermal’, matTag, Fy, E0, b, a1, a2, a3, a4)

This command is the thermal version for 'Steel01'.

matTag integer tag identifying material

(int)
Fy yield strength
(float)
E0 initial elastic tangent
(float)
b (float) strain-hardening ratio (ratio between post-yield tangent and initial elastic tangent)
a1 isotropic hardening parameter, increase of compression yield envelope as proportion of yield
(float) strength after a plastic strain of 𝑎2 * (𝐹𝑦 /𝐸0 ) (optional)
a2 isotropic hardening parameter (see explanation under a1). (optional).
(float)
a3 isotropic hardening parameter, increase of tension yield envelope as proportion of yield strength
(float) after a plastic strain of 𝑎4 * (𝐹𝑦 /𝐸0 ). (optional)
a4 isotropic hardening parameter (see explanation under a3). (optional)
(float)

Concrete Materials

1. Concrete01
2. Concrete02

126 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

3. Concrete04
4. Concrete06
5. Concrete07
6. Concrete01WithSITC
7. ConfinedConcrete01
8. ConcreteD
9. FRPConfinedConcrete
10. FRPConfinedConcrete02
11. ConcreteCM
12. TDConcrete
13. TDConcreteEXP
14. TDConcreteMC10
15. TDConcreteMC10NL

Concrete01

uniaxialMaterial(’Concrete01’, matTag, fpc, epsc0, fpcu, epsU)

This command is used to construct a uniaxial Kent-Scott-Park concrete material object with degraded linear
unloading/reloading stiffness according to the work of Karsan-Jirsa and no tensile strength. (REF: Fedeas).

matTag (int) integer tag identifying material

fpc (float) concrete compressive strength at 28 days (compression is negative)
epsc0 (float) concrete strain at maximum strength
fpcu (float) concrete crushing strength
epsU (float) concrete strain at crushing strength

Note:
1. Compressive concrete parameters should be input as negative values (if input as positive, they will be converted
to negative internally).
2. The initial slope for this model is (2*fpc/epsc0)

See also:
Notes

Concrete02

uniaxialMaterial(’Concrete02’, matTag, fpc, epsc0, fpcu, epsU, lambda, ft, Ets)

This command is used to construct a uniaxial Kent-Scott-Park concrete material object with degraded linear
unloading/reloading stiffness according to the work of Karsan-Jirsa and no tensile strength. (REF: Fedeas).

1.4. Model Commands 127

OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

fpc (float) concrete compressive strength at 28 days (compression is negative)
epsc0 (float) concrete strain at maximum strength
fpcu (float) concrete crushing strength
epsU (float) concrete strain at crushing strength
lambda (float) ratio between unloading slope at $epscu and initial slope
ft (float) tensile strength
Ets (float) tension softening stiffness (absolute value) (slope of the linear tension softening branch)

Note:
1. Compressive concrete parameters should be input as negative values (if input as positive, they will be converted
to negative internally).
2. The initial slope for this model is (2*fpc/epsc0)

See also:
Notes

Concrete04

uniaxialMaterial(’Concrete04’, matTag, fc, epsc, epscu, Ec, fct, et, beta)

This command is used to construct a uniaxial Popovics concrete material object with degraded linear unload-
ing/reloading stiffness according to the work of Karsan-Jirsa and tensile strength with exponential decay.

matTag integer tag identifying material

(int)
fc (float) floating point values defining concrete compressive strength at 28 days (compression is nega-
tive)
epsc floating point values defining concrete strain at maximum strength
(float)
epscu floating point values defining concrete strain at crushing strength
(float)
Ec (float) floating point values defining initial stiffness
fct (float) floating point value defining the maximum tensile strength of concrete (optional)
et (float) floating point value defining ultimate tensile strain of concrete (optional)
beta loating point value defining the exponential curve parameter to define the residual stress (as a
(float) factor of ft) at etu

Note:
1. Compressive concrete parameters should be input as negative values.
2. The envelope of the compressive stress-strain response is defined using the model proposed by Popovics (1973).
If the user defines 𝐸𝑐 = 57000*𝑠𝑞𝑟𝑡(|𝑓 𝑐𝑐|) (in psi)’ then the envelope curve is identical to proposed by Mander
et al. (1988).
3. Model Characteristic: For loading in compression, the envelope to the stress-strain curve follows the model
proposed by Popovics (1973) until the concrete crushing strength is achieved and also for strains beyond that
corresponding to the crushing strength. For unloading and reloading in compression, the Karsan-Jirsa model

128 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(1969) is used to determine the slope of the curve. For tensile loading, an exponential curve is used to define the
envelope to the stress-strain curve. For unloading and reloading in tensile, the secant stiffness is used to define
the path.

See also:
Notes

Concrete06

uniaxialMaterial(’Concrete06’, matTag, fc, e0, n, k, alpha1, fcr, ecr, b, alpha2)

This command is used to construct a uniaxial concrete material object with tensile strength, nonlinear tension
stiffening and compressive behavior based on Thorenfeldt curve.

matTag (int) integer tag identifying material

fc (float) concrete compressive strength (compression is negative)
e0 (float) strain at compressive strength
n (float) compressive shape factor
k (float) post-peak compressive shape factor
alpha1 (float) 𝛼1 parameter for compressive plastic strain definition
fcr (float) tensile strength
ecr (float) tensile strain at peak stress (fcr)
b (float) exponent of the tension stiffening curve
alpha2 (float) 𝛼2 parameter for tensile plastic strain definition

Note:
1. Compressive concrete parameters should be input as negative values.

See also:
Notes

Concrete07

uniaxialMaterial(’Concrete07’, matTag, fc, epsc, Ec, ft, et, xp, xn, r)

Concrete07 is an implementation of Chang & Mander’s 1994 concrete model with simplified unloading and
reloading curves. Additionally the tension envelope shift with respect to the origin proposed by Chang and
Mander has been removed. The model requires eight input parameters to define the monotonic envelope of
confined and unconfined concrete in the following form:

1.4. Model Commands 129

OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
fc (float) concrete compressive strength (compression is negative)
epsc (float) concrete strain at maximum compressive strength
Ec (float) Initial Elastic modulus of the concrete
ft (float) tensile strength of concrete (tension is positive)
et (float) tensile strain at max tensile strength of concrete
xp (float) Non-dimensional term that defines the strain at which the straight line descent begins in
tension
xn (float) Non-dimensional term that defines the strain at which the straight line descent begins in
compression
r (float) Parameter that controls the nonlinear descending branch

See also:
Notes

Concrete01WithSITC

uniaxialMaterial(’Concrete01WithSITC’, matTag, fpc, epsc0, fpcu, epsU, endStrainSITC=0.01)

This command is used to construct a modified uniaxial Kent-Scott-Park concrete material object with degraded
linear unloading/reloading stiffness according to the work of Karsan-Jirsa and no tensile strength. The modifi-
cation is to model the effect of Stuff In The Cracks (SITC).

matTag (int) integer tag identifying material

fpc (float) concrete compressive strength at 28 days (compression is negative)
epsc0 (float) concrete strain at maximum strength
fpcu (float) concrete crushing strength
epsU (float) concrete strain at crushing strength
endStrainSITC (float) optional, default = 0.03

Note:
1. Compressive concrete parameters should be input as negative values (if input as positive, they will be converted
to negative internally).
2. The initial slope for this model is (2*fpc/epsc0)

See also:
Notes

ConfinedConcrete01

uniaxialMaterial(’ConfinedConcrete01’, matTag, secType, fpc, Ec, epscu_type, epscu_val, nu, L1,

L2, L3, phis, S, fyh, Es0, haRatio, mu, phiLon, ’-internal’, *internalArgs, ’-wrap’,
*wrapArgs, ’-gravel’, ’-silica’, ’-tol’, tol, ’-maxNumIter’, maxNumIter, ’-epscuLimit’,
epscuLimit, ’-stRatio’, stRatio)

130 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

secType (str) tag for the transverse reinforcement configuration.
see image below.
• 'S1' square section with S1 type of trans-
verse reinforcement with or without external
FRP wrapping
• 'S2' square section with S2 type of trans-
verse reinforcement with or without external
FRP wrapping
• 'S3' square section with S3 type of trans-
verse reinforcement with or without external
FRP wrapping
• 'S4a' square section with S4a type of trans-
verse reinforcement with or without external
FRP wrapping
• 'S4b' square section with S4b type of trans-
verse reinforcement with or without external
FRP wrapping
• 'S5' square section with S5 type of trans-
verse reinforcement with or without external
FRP wrapping
• 'C' circular section with or without external
FRP wrapping
• 'R' rectangular section with or without exter-
nal FRP wrapping.

fpc (float) unconfined cylindrical strength of concrete speci-

men.
Ec (float) initial elastic modulus of unconfined concrete.
epscu_type (str) Method to define confined concrete ultimate strain
- -epscu then value is confined concrete ultimate
strain, - -gamma then value is the ratio of the
strength corresponding to ultimate strain to the peak
strength of the confined concrete stress-strain curve.
If gamma cannot be achieved in the range [0, ep-
scuLimit] then epscuLimit (optional, default: 0.05)
will be assumed as ultimate strain.
epscu_val (float) Value for the definition of the concrete ultimate strain
nu (str) or (list) Definition for Poisson’s Ratio. - ['-nu',
<value of Poisson's ratio>] -
'-varub' Poisson’s ratio is defined as a function
of axial strain by means of the expression proposed
by Braga et al. (2006) with the upper bound equal
to 0.5 -'-varnoub' Poisson’s ratio is defined as a
function of axial strain by means of the expression
proposed by Braga et al. (2006) without any upper
bound.
L1 (float) length/diameter of square/circular core section mea-
sured respect to the hoop center line.
L2 (float) additional dimensions when multiple hoops are be-
ing used.
L3 (float) additional dimensions when multiple hoops are be-
ing used.
phis (float) hoop diameter. If section arrangement has multiple
hoops it refers to the external hoop.
1.4. Model Commands
S (float) hoop spacing. 131
fyh (float) yielding strength of the hoop steel.
Es0 (float) elastic modulus of the hoop steel.
haRatio (float) hardening ratio of the hoop steel.
OpenSeesPy Documentation, Release 3.5.1.3

See also:
Notes

ConcreteD

uniaxialMaterial(’ConcreteD’, matTag, fc, epsc, ft, epst, Ec, alphac, alphat, cesp=0.25, etap=1.15)
This command is used to construct a concrete material based on the Chinese design code.

matTag (int) integer tag identifying material

fc (float) concrete compressive strength
epsc (float) concrete strain at corresponding to compressive strength
ft (float) concrete tensile strength
epst (float) concrete strain at corresponding to tensile strength
Ec (float) concrete initial Elastic modulus
alphac (float) compressive descending parameter
alphat (float) tensile descending parameter
cesp (float) plastic parameter, recommended values: 0.2~0.3
etap (float) plastic parameter, recommended values: 1.0~1.3

Note:
1. Concrete compressive strength and the corresponding strain should be input as negative values.
2. The value fc/epsc and ft/epst should be smaller than Ec.

See also:

132 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Notes

FRPConfinedConcrete

uniaxialMaterial(’FRPConfinedConcrete’, matTag, fpc1, fpc2, epsc0, D, c, Ej, Sj, tj, eju, S, fyl, fyh,
dlong, dtrans, Es, nu0, k, useBuck)
This command is used to construct a uniaxial Megalooikonomou-Monti-Santini concrete material object with
degraded linear unloading/reloading stiffness according to the work of Karsan-Jirsa and no tensile strength.

matTag integer tag identifying material

(int)
fpc1 (float) concrete core compressive strength.
fpc2 (float) concrete cover compressive strength.
epsc0 strain corresponding to unconfined concrete strength.
(float)
D (float) diameter of the circular section.
c (float) dimension of concrete cover (until the outer edge of steel stirrups)
Ej (float) elastic modulus of the fiber reinforced polymer (FRP) jacket.
Sj (float) clear spacing of the FRP strips - zero if FRP jacket is continuous.
tj (float) total thickness of the FRP jacket.
eju (float) rupture strain of the FRP jacket from tensile coupons.
S (float) spacing of the steel spiral/stirrups.
fyl (float) yielding strength of longitudinal steel bars.
fyh (float) yielding strength of the steel spiral/stirrups.
dlong diameter of the longitudinal bars of the circular section.
(float)
dtrans diameter of the steel spiral/stirrups.
(float)
Es (float) elastic modulus of steel.
nu0 (float) initial Poisson’s coefficient for concrete.
k (float) reduction factor for the rupture strain of the FRP jacket, recommended values 0.5-0.8.
useBuck FRP jacket failure criterion due to buckling of longitudinal compressive steel bars (0 = not
(float) include it, 1= to include it).

Note: #.IMPORTANT: The units of the input parameters should be in MPa, N, mm. #.Concrete compressive strengths
and the corresponding strain should be input as positive values. #.When rupture of FRP jacket occurs due to dilation
of concrete (lateral concrete strain exceeding reduced rupture strain of FRP jacket), the analysis is not terminated.
Only a message “FRP Rupture” is plotted on the screen. #.When $useBuck input parameter is on (equal to 1) and
the model’s longitudinal steel buckling conditions are fulfilled, a message “Initiation of Buckling of Long.Bar under
Compression” is plotted on the screen. #.When rupture of FRP jacket occurs due to its interaction with buckled
longitudinal compressive steel bars, the analysis is not terminated. Only a message “FRP Rupture due to Buckling of
Long.Bar under compression” is plotted on the screen.

See also:
Notes

FRPConfinedConcrete02

DEVELOPED AND IMPLEMENTED BY:

1.4. Model Commands 133

OpenSeesPy Documentation, Release 3.5.1.3

Jin-Yu LU, Southeast University, Nanjing, China

Guan LIN ([email protected]), Hong Kong Polytechnic University, Hong Kong, China.

uniaxialMaterial(’FRPConfinedConcrete02’, matTag, fc0, Ec, ec0, <’-JacketC’, tfrp, Efrp, erup, R>,
<’-Ultimate’, fcu, ecu>, ft, Ets, Unit)
Figure 1 Hysteretic Stress-Strain Relation

This command is used to construct a uniaxial hysteretic stress-strain model for fiber-reinforced polymer (FRP)-
confined concrete. The envelope compressive stress-strain response is described by a parabolic first portion and
a linear second portion with smooth connection between them (Figure 1). The hysteretic rules of compression
are based on Lam and Teng’s (2009) model. The cyclic linear tension model of Yassin (1994) for unconfined
concrete (as adopted in Concrete02) is used with slight modifications to describe the tensile behavior of FRP-
confined concrete (Teng et al. 2015).

matTag (int) integer tag identifying material

fc0 (float) compressive strength of unconfined concrete (compression is negative)
Ec (float) elastic modulus of unconfined concrete (=4730(-$fc0(MPa)))
ec0 (float) axial strain corresponding to unconfined concrete strength ( 0.002)
-JacketC (str) input parameters of the FRP jacket in a circular section
tfrp (float) thickness of an FRP jacket
Efrp (float) tensile elastic modulus of an FRP jacket
erup (float) hoop rupture strain of an FRP jacket
R (float) radius of circular column section
-Ultimate (str) input ultimate stress/strain directly
fcu (float) ultimate stress of FRP-confined concrete ($fcu $fc0)
ecu (float) ultimate strain of FRP-confined concrete
ft (float) tensile strength of unconfined concrete (=0.632(-$fc0(MPa)))
Ets (float) stiffness of tensile softening ( 0.05 Ec)
Unit (float) unit indicator, Unit = 1 for SI Metric Units; Unit = 0 for US Customary Units

Note:
1. Compressive concrete parameters should be input as negative values.
2. The users are required to input either the FRP jacket properties in an FRP-confined circular column (<-JacketC>)

134 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

or directly input the ultimate point (𝜖cu, fcu) (<-Ultimate>). If <-JacketC> is used, the ultimate stress and strain
are automatically calculated based on Teng et al.’s (2009) model which is a refined version of Lam and Teng’s
(2003) stress-strain model for FRP-confined concrete in circular columns. If <-Ultimate> is used, the ultimate
stress and strain can be calculated by the users in advance based on other stress-strain models of FRP-confined
concrete and thus can be used for other cross section shapes (e.g., square, rectangular, or elliptical). If none
of them is specified, a stress-strain curve (parabola + horizontal linear curve) for unconfined concrete will be
defined (Figure 1). Both <-JacketC> and <-Ultimate> adopt the envelope compressive stress-strain curve with a
parabolic first portion and a linear second portion.
3. Unit indicator: $Unit = 1 for SI Metric Units (e.g., N, mm, MPa); $Unit = 0 for US Customary Units (e.g., kip,
in, sec, ksi).
Calibration:
1. The implemented new material has been calibrated using a simple-supported Force-Based Beam-Column
element subjected to axial load only (http://opensees.berkeley.edu/wiki/index.php/Calibration_of_Maxwell_
Material). The output stress-strain responses were compared with the desired curves defined by the input pa-
rameters.
Examples:
1. Example 1: Pin-ended FRP-confined reinforced concrete (RC) columns
Figure 2 Simulation of pin-ended FRP-confined RC column

1. The first example is a pin-ended FRP-confined circular RC column subjected to eccentric com-
pression (load eccentricity = 20 mm) at both ends tested by Bisby and Ranger (2010) (Figure 2).
Due to the symmetry in geometry and loading, only half of the column needs to be modelled. In
this case, three forceBeamColumn elements each with 5 integration points were used for the half
column. The FRPConfinedConcrete02 model was used to describe the stress-strain behavior of
FRP-confined concrete. Either <-JacketC> or <-Ultimate> can be used. If the former is used,
the properties of the FRP jacket need to be input; if the latter is used, the ultimate stress and
strain need to be calculated by the users and input directly. The eccentric loading is applied with
a combined axial load and bending moment at each end node. An increasing vertical displace-
ment is applied to the top node of the column model. The analysis terminated until the ultimate
axial strain of FRP-confined concrete was reached by the extreme compression concrete fiber at
the mid-height (equivalent to FRP rupture). SI Metric Unit (e.g., N, mm, MPa) is used in the
script of this example ($Unit = 1).
2. Figure 3 shows the comparison of axial load-lateral displacement curve between the test results
and the theoretical results. Figure 4 shows the variation of column slenderness ratio (l/D) on

1.4. Model Commands 135

OpenSeesPy Documentation, Release 3.5.1.3

the axial load-lateral displacement response of the column. Please refer to Lin (2016) for more
details about the modeling.
Figure 3 Experimental results vs theoretical results

Figure 4 Parametric study (effect of column slenderness ratio)

2. Example 2: Cantilever column subjected to constant axial compression and cyclic lateral loading
Figure 5 Simulation of columns under cyclic latera loading

136 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1. The second example is a cantilever FRP-confined circular RC column subjected to constant axial
compression and cyclic lateral loading (Column C5 tested by Saadatmanesh et al. 1997). The
US Customary Units (e.g., kip, in, sec, ksi) were used in this example. The twenty-five (25)-in.-
height region (potential plastic hinge region) above the footing of the column was wrapped with
an FRP jacket; the remaining portion of the column with a height of 71 in. was conventional
RC section without FRP jacketing. The column was modelled using two forceBeamColumn
elements to cater for the variation of section characteristic along the column height. A zero
length section element at the column-footing interface was used to simulate fixed-end rotations
due to the strain penetration of longitudinal steel bars (Figure 5) (Lin et al. 2012). The bond-slip
model of Zhao and Sritharan (2007) (Bond_SP01) was used to depict the bar stress-slip response.
In addition, another zero length section element was used at the column-footing interface to
consider the possible rotations of the footing (Teng et al. 2015). The rotation stiffness of the zero
length section element was adjusted to achieve close matching between the test response and the
predicted response during the initial stage of loading. This zero length section element was found
to have little effect on the ultimate displacement of the column (Teng et al. 2015). Moreover, the
inclination of axial load in the column test needs to be accounted for when comparing predicted
results with test results (Teng et al. 2015). Figure 6 shows the comparison of lateral load-lateral
displacement curve between the test results and the theoretical results.
References:
1. Bisby, L. and Ranger, M. (2010). “Axial-flexural interaction in circular FRP-confined reinforced concrete
columns”, Construction and Building Materials, Vol. 24, No. 9, pp. 1672-1681.
2. Lam, L. and Teng, J.G. (2003). “Design-oriented stress-strain model for FRP-confined concrete”, Construction
and Building Materials, Vol. 17, No. 6, pp. 471-489.
3. Lam, L. and Teng, J.G. (2009). “Stress-strain model for FRP-confined concrete under cyclic axial compression”,
Engineering Structures, Vol. 31, No. 2, pp. 308-321.
4. Lin, G. (2016). Seismic Performance of FRP-confined RC Columns: Stress-Strain Models and Numerical
Simulation, Ph.D. thesis, Department of Civil and Environmental Engineering, The Hong Kong Polytechnic
University, Hong Kong, China.
5. Lin, G. and Teng, J.G. (2015). “Numerical simulation of cyclic/seismic lateral response of square RC columns
confined with fibre-reinforced polymer jackets”, Proceedings, Second International Conference on Performance-
based and Life-cycle Structural Engineering (PLSE 2015), pp. 481-489 (http://plse2015.org/cms/USB/pdf/
full-paper_7408.pdf).
6. Lin, G., Teng, J.G. and Lam, L. (2012). “Numerical simulation of FRP-jacketed RC columns under cyclic

1.4. Model Commands 137

OpenSeesPy Documentation, Release 3.5.1.3

loading: modeling of the strain penetration effect”, First International Conference on Performance-based and
Life-cycle Structural Engineering (PLSE2012), December 5-7, Hong Kong, China.
7. Saadatmanesh, H., Ehsani, M. and Jin, L. (1997). “Seismic retrofitting of rectangular bridge columns with
composite straps”, Earthquake Spectra, Vol. 13, No. 2, pp. 281-304.
8. Teng, J.G., Lam, L., Lin, G., Lu, J.Y. and Xiao, Q.G. (2015). “Numerical Simulation of FRP-Jacketed RC
Columns Subjected to Cyclic and Seismic Loading”, Journal of Composites for Construction, ASCE, Vol. 20,
No. 1, pp. 04015021.
9. Yassin, M.H.M. (1994). Nonlinear Analysis of Prestressed Concrete Structures under Monotonic and Cyclic
Loads, Ph.D. thesis, University of California at Berkeley, California, USA.
10. Zhao, J. and Sritharan, S. (2007). “Modeling of strain penetration effects in fiber-based analysis of reinforced
concrete structuresconcrete structures”, ACI Structural Journal, Vol. 104, No. 2, pp. 133-141.

ConcreteCM

uniaxialMaterial(’ConcreteCM’, matTag, fpcc, epcc, Ec, rc, xcrn, ft, et, rt, xcrp, mon, ’-GapClose’,
GapClose=0)
This command is used to construct a uniaxialMaterial ConcreteCM (Kolozvari et al., 2015), which is a uniaxial
hysteretic constitutive model for concrete developed by Chang and Mander (1994).

matTag integer tag identifying material

(int)
fpcc Compressive strength (𝑓𝑐′ )
(float)
epcc Strain at compressive strength (𝜖′𝑐 )
(float)
Ec (float)
Initial tangent modulus (𝐸𝑐 )
rc (float)
Shape parameter in Tsai’s equation defined for compression (𝑟𝑐 )
xcrn Non-dimensional critical strain on compression envelope (𝜖− 𝑐𝑟 , where the envelope curve starts
(float) following a straight line)
ft (float)
Tensile strength (𝑓𝑡 )
et (float)
Strain at tensile strength (𝜖𝑡 )
rt (float)
Shape parameter in Tsai’s equation defined for tension (𝑟𝑡 )
xcrp Non-dimensional critical strain on tension envelope (𝜖+ 𝑐𝑟 , where the envelope curve starts fol-
(float) lowing a straight line - large value [e.g., 10000] recommended when tension stiffening is con-
sidered)
mon optional, monotonic stress-strain relationship only: mon=1 (invoked in FSAM only), mon=0
(no impact since monotonic)
optional, denote next parameter is GapClose
'-GapClose'
(str)
GapClose optional, GapClose = 0, less gradual gap closure (default); GapClose = 1, more gradual gap
(float) closure

See also:
Notes

138 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

TDConcrete

uniaxialMaterial(’TDConcrete’, matTag, fc, fct, Ec, beta, tD, epsshu, psish, Tcr, phiu, psicr1, psicr2,
tcast)
This command is used to construct a uniaxial time-dependent concrete material object with linear behavior in
compression, nonlinear behavior in tension (REF: Tamai et al., 1988) and creep and shrinkage according to ACI
209R-92.

matTag (int) integer tag identifying material

fc (float) concrete compressive strength (compression is negative)
fct (float) concrete tensile strength (tension is positive)
Ec (float) concrete modulus of elasticity
beta (float) tension softening parameter (tension softening exponent)
tD (float) analysis time at initiation of drying (in days)
epsshu (float) ultimate shrinkage strain as per ACI 209R-92 (shrinkage is negative)
psish (float) fitting parameter of the shrinkage time evolution function as per ACI 209R-92
Tcr (float) creep model age (in days)
phiu (float) ultimate creep coefficient as per ACI 209R-92
psicr1 (float) fitting parameter of the creep time evolution function as per ACI 209R-92
psicr2 (float) fitting parameter of the creep time evolution function as per ACI 209R-92
tcast (float) analysis time corresponding to concrete casting (in days; minimum value 2.0)

Note:
1. Compressive concrete parameters should be input as negative values (if input as positive, they will be converted
to negative internally).
2. Shrinkage concrete parameters should be input as negative values (if input as positive, they will be converted to
negative internally).

See also:
Detailed descriptions of the model and its implementation can be found in the following: (1) Knaack, A.M., Kurama,
Y.C. 2018. Modeling Time-Dependent Deformations: Application for Reinforced Concrete Beams with Recycled
Concrete Aggregates. ACI Structural J. 115, 175-190. doi:10.14359/51701153 (2) Knaack, A.M., 2013. Sustain-
able concrete structures using recycled concrete aggregate: short-term and long-term behavior considering material
variability. PhD Dissertation, Civil and Environmental Engineering and Earth Sciences, University of Notre Dame,
Notre Dame, Indiana, USA, 680 pp A manual describing the use of the model and sample files can be found at:
https://data.mendeley.com/datasets/z4gxnhchky/5

TDConcreteEXP

uniaxialMaterial(’TDConcreteEXP’, matTag, fc, fct, Ec, beta, tD, epsshu, psish, Tcr, epscru, sigCr,
psicr1, psicr2, tcast)
This command is used to construct a uniaxial time-dependent concrete material object with linear behavior in
compression, nonlinear behavior in tension (REF: Tamai et al., 1988) and creep and shrinkage according to ACI
209R-92.

1.4. Model Commands 139

OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

fc (float) concrete compressive strength (compression is negative)
fct (float) concrete tensile strength (tension is positive)
Ec (float) concrete modulus of elasticity
beta (float) tension softening parameter (tension softening exponent)
tD (float) analysis time at initiation of drying (in days)
epsshu ultimate shrinkage strain as per ACI 209R-92 (shrinkage is negative)
(float)
psish (float) fitting parameter of the shrinkage time evolution function as per ACI 209R-92
Tcr (float) creep model age (in days)
epscru ultimate creep strain (e.g., taken from experimental measurements)
(float)
sigCr (float) concrete compressive stress (input as negative) associated with $epscru (e.g., experimen-
tally applied)
psicr1 fitting parameter of the creep time evolution function as per ACI 209R-92
(float)
psicr2 fitting parameter of the creep time evolution function as per ACI 209R-92
(float)
tcast (float) analysis time corresponding to concrete casting (in days; minimum value 2.0)

Note:
1. Compressive concrete parameters should be input as negative values (if input as positive, they will be converted
to negative internally).
2. Shrinkage concrete parameters should be input as negative values (if input as positive, they will be converted to
negative internally).

See also:
Detailed descriptions of the model and its implementation can be found in the following: (1) Knaack, A.M., Kurama,
Y.C. 2018. Modeling Time-Dependent Deformations: Application for Reinforced Concrete Beams with Recycled
Concrete Aggregates. ACI Structural J. 115, 175-190. doi:10.14359/51701153 (2) Knaack, A.M., 2013. Sustain-
able concrete structures using recycled concrete aggregate: short-term and long-term behavior considering material
variability. PhD Dissertation, Civil and Environmental Engineering and Earth Sciences, University of Notre Dame,
Notre Dame, Indiana, USA, 680 pp A manual describing the use of the model and sample files can be found at:
https://data.mendeley.com/datasets/z4gxnhchky/5

TDConcreteMC10

uniaxialMaterial(’TDConcreteMC10’, matTag, fc, fct, Ec, Ecm, beta, tD, epsba, epsbb, epsda, epsdb,
phiba, phibb, phida, phidb, tcast, cem)
This command is used to construct a uniaxial time-dependent concrete material object with linear behavior in
compression, nonlinear behavior in tension (REF: Tamai et al., 1988) and creep and shrinkage according to fib
Model Code 2010.

140 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
fc (float) concrete compressive strength (compression is negative)
fct (float) concrete tensile strength (tension is positive)
Ec (float) concrete modulus of elasticity at loading age
Ecm (float) concrete modulus of elasticity at 28 days
beta tension softening parameter (tension softening exponent)
(float)
tD (float) analysis time at initiation of drying (in days)
epsba ultimate basic shrinkage strain (input as negative) as per fib Model Code 2010
(float)
epsbb fitting parameter of the basic shrinkage time evolution function as per fib Model Code 2010
(float)
epsda product of ultimate drying shrinkage strain and relative humidity function as per fib Model
(float) Code 2010
epsdb fitting parameter of the basic shrinkage time evolution function as per fib Model Code 2010
(float)
phiba parameter for the effect of compressive strength on basic creep as per fib Model Code 2010
(float)
phibb fitting parameter of the basic creep time evolution function as per fib Model Code 2010
(float)
phida product of the effect of compressive strength and relative humidity on drying creep as per fib
(float) Model Code 2010
phidb fitting parameter of the drying creep time evolution function as per fib Model Code 2010
(float)
tcast analysis time corresponding to concrete casting (in days; minimum value 2.0)
(float)
cem (float) coefficient dependent on the type of cement as per fib Model Code 2010

Note:
1. Compressive concrete parameters should be input as negative values (if input as positive, they will be converted
to negative internally).
2. Shrinkage concrete parameters should be input as negative values (if input as positive, they will be converted to
negative internally).

See also:
Detailed descriptions of the model and its implementation can be found in the following: (1) Knaack, A.M., Kurama,
Y.C. 2018. Modeling Time-Dependent Deformations: Application for Reinforced Concrete Beams with Recycled
Concrete Aggregates. ACI Structural J. 115, 175-190. doi:10.14359/51701153 (2) Knaack, A.M., 2013. Sustain-
able concrete structures using recycled concrete aggregate: short-term and long-term behavior considering material
variability. PhD Dissertation, Civil and Environmental Engineering and Earth Sciences, University of Notre Dame,
Notre Dame, Indiana, USA, 680 pp A manual describing the use of the model and sample files can be found at:
https://data.mendeley.com/datasets/z4gxnhchky/5

TDConcreteMC10NL

uniaxialMaterial(’TDConcreteMC10NL’, matTag, fc, fcu, epscu, fct, Ec, Ecm, beta, tD, epsba, epsbb,
epsda, epsdb, phiba, phibb, phida, phidb, tcast, cem)
This command is used to construct a uniaxial time-dependent concrete material object with non-linear behavior

1.4. Model Commands 141

OpenSeesPy Documentation, Release 3.5.1.3

in compression (REF: Concrete02), nonlinear behavior in tension (REF: Tamai et al., 1988) and creep and
shrinkage according to fib Model Code 2010.

matTag integer tag identifying material

(int)
fc (float) concrete compressive strength (compression is negative)
fcu (float) concrete crushing strength (compression is negative)
epscu concrete strain at crushing strength (input as negative)
(float)
fct (float) concrete tensile strength (tension is positive)
Ec (float) concrete modulus of elasticity at loading age
Ecm (float) concrete modulus of elasticity at 28 days
beta tension softening parameter (tension softening exponent)
(float)
tD (float) analysis time at initiation of drying (in days)
epsba ultimate basic shrinkage strain (input as negative) as per fib Model Code 2010
(float)
epsbb fitting parameter of the basic shrinkage time evolution function as per fib Model Code 2010
(float)
epsda product of ultimate drying shrinkage strain and relative humidity function as per fib Model
(float) Code 2010
epsdb fitting parameter of the basic shrinkage time evolution function as per fib Model Code 2010
(float)
phiba parameter for the effect of compressive strength on basic creep as per fib Model Code 2010
(float)
phibb fitting parameter of the basic creep time evolution function as per fib Model Code 2010
(float)
phida product of the effect of compressive strength and relative humidity on drying creep as per fib
(float) Model Code 2010
phidb fitting parameter of the drying creep time evolution function as per fib Model Code 2010
(float)
tcast analysis time corresponding to concrete casting (in days; minimum value 2.0)
(float)
cem (float) coefficient dependent on the type of cement as per fib Model Code 2010

Note:
1. Compressive concrete parameters should be input as negative values (if input as positive, they will be converted
to negative internally).
2. Shrinkage concrete parameters should be input as negative values (if input as positive, they will be converted to
negative internally).

See also:
Detailed descriptions of the model and its implementation can be found in the following: (1) Knaack, A.M., Kurama,
Y.C. 2018. Modeling Time-Dependent Deformations: Application for Reinforced Concrete Beams with Recycled
Concrete Aggregates. ACI Structural J. 115, 175-190. doi:10.14359/51701153 (2) Knaack, A.M., 2013. Sustain-
able concrete structures using recycled concrete aggregate: short-term and long-term behavior considering material
variability. PhD Dissertation, Civil and Environmental Engineering and Earth Sciences, University of Notre Dame,
Notre Dame, Indiana, USA, 680 pp A manual describing the use of the model and sample files can be found at:
https://data.mendeley.com/datasets/z4gxnhchky/5

142 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Standard Uniaxial Materials

1. Elastic Uniaxial Material

2. Elastic-Perfectly Plastic Material
3. Elastic-Perfectly Plastic Gap Material
4. Elastic-No Tension Material
5. Parallel Material
6. Series Material

Elastic Uniaxial Material

uniaxialMaterial(’Elastic’, matTag, E, eta=0.0, Eneg=E)

This command is used to construct an elastic uniaxial material object.

matTag (int) integer tag identifying material

E (float) tangent
eta (float) damping tangent (optional, default=0.0)
Eneg (float) tangent in compression (optional, default=E)

See also:
Notes

Elastic-Perfectly Plastic Material

uniaxialMaterial(’ElasticPP’, matTag, E, epsyP, epsyN=epsyP, eps0=0.0)

This command is used to construct an elastic perfectly-plastic uniaxial material object.

matTag integer tag identifying material

(int)
E (float) tangent
epsyP strain or deformation at which material reaches plastic state in tension
(float)
epsyN strain or deformation at which material reaches plastic state in compression. (optional, default
(float) is tension value)
eps0 initial strain (optional, default: zero)
(float)

See also:
Notes

Elastic-Perfectly Plastic Gap Material

uniaxialMaterial(’ElasticPPGap’, matTag, E, Fy, gap, eta=0.0, damage=’noDamage’)

This command is used to construct an elastic perfectly-plastic gap uniaxial material object.

1.4. Model Commands 143

OpenSeesPy Documentation, Release 3.5.1.3

matTaginteger tag identifying material

(int)
E tangent
(float)
Fy stress or force at which material reaches plastic state
(float)
gap initial gap (strain or deformation)
(float)
eta hardening ratio (=Eh/E), which can be negative
(float)
damagean optional string to specify whether to accumulate damage or not in the material. With the default
(str) string, 'noDamage' the gap material will re-center on load reversal. If the string 'damage' is
provided this recentering will not occur and gap will grow.

See also:
Notes

Elastic-No Tension Material

uniaxialMaterial(’ENT’, matTag, E)
This command is used to construct a uniaxial elastic-no tension material object.

matTag (int) integer tag identifying material

E (float) tangent

See also:
Notes

Parallel Material

uniaxialMaterial(’Parallel’, matTag, *MatTags, ’-factors’, *factorArgs)

This command is used to construct a parallel material object made up of an arbitrary number of previously-
constructed UniaxialMaterial objects.

matTag (int) integer tag identifying material

MatTags (list identification tags of materials making up the material model
(int))
factorArgs factors to create a linear combination of the specified materials. Factors can be negative
(list (float)) to subtract one material from an other. (optional, default = 1.0)

See also:
Notes

Series Material

uniaxialMaterial(’Series’, matTag, *matTags)

This command is used to construct a series material object made up of an arbitrary number of previously-
constructed UniaxialMaterial objects.

144 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

matTags (list (int)) identification tags of materials making up the material model

See also:
Notes

PyTzQz uniaxial materials for p-y, t-z and q-z elements for modeling soil-structure interaction
through the piles in a structural foundation

1. PySimple1 Material
2. TzSimple1 Material
3. QzSimple1 Material
4. PyLiq1 Material
5. TzLiq1 Material
6. QzLiq1 Material

PySimple1 Material

uniaxialMaterial(’PySimple1’, matTag, soilType, pult, Y50, Cd, c=0.0)

This command is used to construct a PySimple1 uniaxial material object.

matTag integer tag identifying material

(int)
soilTypesoilType = 1 Backbone of p-y curve approximates Matlock (1970) soft clay relation.
(int) soilType = 2 Backbone of p-y curve approximates API (1993) sand relation.
pult Ultimate capacity of the p-y material. Note that “p” or “pult” are distributed loads [force per length
(float) of pile] in common design equations, but are both loads for this uniaxialMaterial [i.e., distributed
load times the tributary length of the pile].
Y50 Displacement at which 50% of pult is mobilized in monotonic loading.
(float)
Cd Variable that sets the drag resistance within a fully-mobilized gap as Cd*pult.
(float)
c The viscous damping term (dashpot) on the far-field (elastic) component of the displacement rate
(float) (velocity). (optional Default = 0.0). Nonzero c values are used to represent radiation damping
effects

See also:
Notes

TzSimple1 Material

uniaxialMaterial(’TzSimple1’, matTag, soilType, tult, z50, c=0.0)

This command is used to construct a TzSimple1 uniaxial material object.

1.4. Model Commands 145

OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
soilType soilType = 1 Backbone of t-z curve approximates Reese and O’Neill (1987).
(int) soilType = 2 Backbone of t-z curve approximates Mosher (1984) relation.
tult Ultimate capacity of the t-z material. SEE NOTE 1.
(float)
z50 Displacement at which 50% of tult is mobilized in monotonic loading.
(float)
c (float) The viscous damping term (dashpot) on the far-field (elastic) component of the displacement
rate (velocity). (optional Default = 0.0). See NOTE 2.

Note:
1. The argument tult is the ultimate capacity of the t-z material. Note that t or tult are shear stresses [force
per unit area of pile surface] in common design equations, but are both loads for this uniaxialMaterial [i.e., shear
stress times the tributary area of the pile].
2. Nonzero c values are used to represent radiation damping effects

See also:
Notes

QzSimple1 Material

uniaxialMaterial(’QzSimple1’, matTag, qzType, qult, Z50, suction=0.0, c=0.0)

This command is used to construct a QzSimple1 uniaxial material object.

matTag integer tag identifying material

(int)
qzType qzType = 1 Backbone of q-z curve approximates Reese and O’Neill’s (1987) relation for drilled
(int) shafts in clay.
qzType = 2 Backbone of q-z curve approximates Vijayvergiya’s (1977) relation for piles in sand.
qult Ultimate capacity of the q-z material. SEE NOTE 1.
(float)
Z50 Displacement at which 50% of qult is mobilized in monotonic loading. SEE NOTE 2.
(float)
suction Uplift resistance is equal to suction*qult. Default = 0.0. The value of suction must be 0.0 to 0.1.
(float)
c (float) The viscous damping term (dashpot) on the far-field (elastic) component of the displacement rate
(velocity). Default = 0.0. Nonzero c values are used to represent radiation damping effects.*

Note:
1. qult: Ultimate capacity of the q-z material. Note that q1 or qult are stresses [force per unit area of pile tip]
in common design equations, but are both loads for this uniaxialMaterial [i.e., stress times tip area].
2. Nonzero c values are used to represent radiation damping effects
3. Z50: Displacement at which 50% of qult is mobilized in monotonic loading. Note that Vijayvergiya’s rela-
tion (qzType=2) refers to a “critical” displacement (zcrit) at which qult is fully mobilized, and that the
corresponding z50 would be 0. 125zcrit.

146 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

4. optional args suction and c must either both be omitted or both provided.

See also:
Notes

PyLiq1 Material

This command constructs a uniaxial p-y material that incorporates liquefaction effects. This p y material is
used with a zeroLength element to connect a pile (beam-column element) to a 2 D plane-strain FE mesh or
displacement boundary condition. The p-y material obtains the average mean effective stress (which de-
creases with increasing excess pore pressure) either from two specified soil elements, or from a time series.
Currently, the implementation requires that the specified soil elements consist of FluidSolidPorousMateri-
als in FourNodeQuad elements, or PressureDependMultiYield or PressureDependMultiYield02 materials
in FourNodeQuadUP or NineFourQuadUP elements. There are two possible forms:
uniaxialMaterial(’PyLiq1’, matTag, soilType, pult, Y50, Cd, c, pRes, ele1, ele2)
uniaxialMaterial(’PyLiq1’, matTag, soilType, pult, Y50, Cd, c, pRes, ’-timeSeries’, timeSeriesTag)

matTag integer tag identifying material

(int)
soilType soilType = 1 Backbone of p-y curve approximates Matlock (1970) soft clay relation. soilType =
(int) 2 Backbone of p-y curve approximates API (1993) sand relation.
pult Ultimate capacity of the p-y material. Note that “p” or “pult” are distributed loads [force per
(float) length of pile] in common design equations, but are both loads for this uniaxialMaterial [i.e.,
distributed load times the tributary length of the pile].
Y50 Displacement at which 50% of pult is mobilized in monotonic loading.
(float)
Cd (float) Variable that sets the drag resistance within a fully-mobilized gap as Cd*pult.
c (float) The viscous damping term (dashpot) on the far-field (elastic) component of the displacement rate
(velocity). (optional Default = 0.0). Nonzero c values are used to represent radiation damping
effects
pRes sets the minimum (or residual) peak resistance that the material retains as the adjacent solid soil
(float) elements liquefy
ele1 are the eleTag (element numbers) for the two solid elements from which PyLiq1 will obtain
ele2 mean effective stresses and excess pore pressures
(int)
Alternatively, mean effective stress can be supplied by a time series by specifying the text string
timeSeriesTag
(int) '-timeSeries' and the tag of the series seriesTag.

Note:
1. The argument pult is the ultimate capacity of the p-y material. Note that p or pult are lateral stresses [force
per unit length of pile] in common design equations, but are both loads for this uniaxialMaterial [i.e., distributed
load times the tributary length of the pile].
2. Nonzero c values are used to represent radiation damping effects
3. To model the effects of liquefaction with PyLiq1, it is necessary to use the updateMaterialStage com-
mand. When material stage is 0 (which is the default value), the PyLiq1 behavior will be independent of any
pore pressure in the specified solidElem’s. When material stage is set to 1, the behviour of PyLiq1 will depend
on the mean effective stresses (and hence excess pore pressures) in the solidElem’s.

1.4. Model Commands 147

OpenSeesPy Documentation, Release 3.5.1.3

See also:
Notes

TzLiq1 Material

The command constructs a uniaxial t-z material that incorporates liquefaction effects. This t z material
is used with a zeroLength element to connect a pile (beam-column element) to a 2 D plane-strain FE
mesh. The t-z material obtains the average mean effective stress (which decreases with increasing excess
pore pressure) from two specified soil elements. Currently, the implementation requires that the specified
soil elements consist of FluidSolidPorousMaterials in FourNodeQuad elements. There are two possible
forms:
uniaxialMaterial(’TzLiq1’, matTag, tzType, tult, z50, c, ele1, ele2)
uniaxialMaterial(’TzLiq1’, matTag, tzType, tult, z50, c, ’-timeSeries’, timeSeriesTag)

matTag (int) integer tag identifying material

tzType (int) tzType = 1 Backbone of t-z curve approximates Reese and O’Neill (1987). tzType = 2
Backbone of t-z curve approximates Mosher (1984) relation.
tult (float) Ultimate capacity of the t-z material. SEE NOTE 1.
z50 (float) Displacement at which 50% of tult is mobilized in monotonic loading.
c (float) The viscous damping term (dashpot) on the far-field (elastic) component of the displace-
ment rate (velocity).
ele1 ele2 are the eleTag (element numbers) for the two solid elements from which PyLiq1 will obtain
(int) mean effective stresses and excess pore pressures
timeSeriesTagAlternatively, mean effective stress can be supplied by a time series by specifying the text
(int) string '-timeSeries' and the tag of the seriesm seriesTag.

Note:
1. The argument tult is the ultimate capacity of the t-z material. Note that t or tult are shear stresses [force
per unit area of pile surface] in common design equations, but are both loads for this uniaxialMaterial [i.e., shear
stress times the tributary area of the pile].
2. Nonzero c values are used to represent radiation damping effects
3. To model the effects of liquefaction with TzLiq1, it is necessary to use the updateMaterialStage com-
mand. When material stage is 0 (which is the default value), the TzLiq1 behavior will be independent of any
pore pressure in the specified solidElem’s. When material stage is set to 1, the behviour of TzLiq1 will depend
on the mean effective stresses (and hence excess pore pressures) in the solidElem’s.

See also:
Notes

QzLiq1 Material

The command constructs a uniaxial q-z material that incorporates liquefaction effects. This q-z material
is used with a zeroLength element to connect a pile (beam-column element) to a 2 D plane-strain FE
mesh. The q-z material obtains the average mean effective stress (which decreases with increasing excess

148 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

pore pressure) from two specified soil elements or from the mean effective stress provided explicitly as
a timeseries data. Currently, in order to compute the mean effective stress from connected elements, the
implementation requires that the specified soil elements consist of FluidSolidPorousMaterials in FourN-
odeQuad elements. There are two possible forms:
uniaxialMaterial(’QzLiq1’, matTag, soilType, qult, Z50, Cd, c, alpha, ele1, ele2)
uniaxialMaterial(’QzLiq1’, matTag, soilType, qult, Z50, Cd, c, alpha, ’-timeSeries’, timeSeriesTag)

matTag (int) integer tag identifying material

soilType soilType = 1 Backbone of q-z curve approximates Matlock (1970) soft clay relation.
(int) soilType = 2 Backbone of q-z curve approximates API (1993) sand relation.
qult (float) Ultimate capacity of the q-z material.
Z50 (float) Displacement at which 50% of qult is mobilized in monotonic loading.
suction Uplift resistance is equal to suction*qult.
(float) Default=0.0. The value of suction must be 0.0 to 0.1.
c (float) The viscous damping term (optional Default = 0.0).
alpha The constant alpha defines the extent of non-linearity in element’s capacity and stiffness
(float) with excess pore pressure ratio (𝑟𝑢 ).
ele1 ele2 are the eleTag (element numbers) for the two solid elements from which QzLiq1 will obtain
(int) mean effective stresses and excess pore alphasures
timeSeriesTagAlternatively, mean effective stress can be supplied by a time series by specifying the text
(int) string '-timeSeries' and the tag of the series seriesTag.

Note:
1. qult: Ultimate capacity of the q-z material. Note that q or qult are stresses [force per unit area of pile tip] in
common design equations, but are both loads for this uniaxialMaterial [i.e., stress times tip area].
2. Z50: Displacement at which 50% of qult is mobilized in monotonic loading. Note that Vijayvergiya’s rela-
tion (qzType=2) refers to a “critical” displacement (zcrit) at which qult is fully mobilized, and that the
corresponding z50 would be 0. 125zcrit.
3. Nonzero c values are used to represent radiation damping effects
4. To model the effects of liquefaction with QzLiq1, it is necessary to use the updateMaterialStage com-
mand. When material stage is 0 (which is the default value), the QzLiq1 behavior will be independent of
any pore alphasure in the specified solidElem’s. When material stage is set to 1, the behviour of QzLiq1 will
depend on the mean effective stresses (and hence excess pore alphasures) in the solidElem’s.

See also:
Notes

Other Uniaxial Materials

1. Hardening Material
2. CastFuse Material
3. ViscousDamper Material
4. BilinearOilDamper Material
5. Modified Ibarra-Medina-Krawinkler Deterioration Model with Bilinear Hysteretic Response (Bilin Material)

1.4. Model Commands 149

OpenSeesPy Documentation, Release 3.5.1.3

6. Modified Ibarra-Medina-Krawinkler Deterioration Model with Peak-Oriented Hysteretic Response (ModIMK-

PeakOriented Material)
7. Modified Ibarra-Medina-Krawinkler Deterioration Model with Pinched Hysteretic Response (ModIMKPinching
Material)
8. SAWS Material
9. BarSlip Material
10. Bond SP01 - - Strain Penetration Model for Fully Anchored Steel Reinforcing Bars
11. Fatigue Material
12. Impact Material
13. Hyperbolic Gap Material
14. Limit State Material
15. MinMax Material
16. ElasticBilin Material
17. ElasticMultiLinear Material
18. MultiLinear
19. Initial Strain Material
20. Initial Stress Material
21. PathIndependent Material
22. Pinching4 Material
23. Engineered Cementitious Composites Material
24. SelfCentering Material
25. Viscous Material
26. BoucWen Material
27. BWBN Material
28. KikuchiAikenHDR Material
29. KikuchiAikenLRB Material
30. AxialSp Material
31. AxialSpHD Material
32. Pinching Limit State Material
33. CFSWSWP Wood-Sheathed Cold-Formed Steel Shear Wall Panel
34. CFSSSWP Steel-Sheathed Cold-formed Steel Shear Wall Panel
35. Backbone Material

Hardening Material

uniaxialMaterial(’Hardening’, matTag, E, sigmaY, H_iso, H_kin, eta=0.0)

This command is used to construct a uniaxial material object with combined linear kinematic and isotropic
hardening. The model includes optional visco-plasticity using a Perzyna formulation.

150 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

E (float) tangent stiffness
sigmaY (float) yield stress or force
H_iso (float) isotropic hardening Modulus
H_kin (float) kinematic hardening Modulus
eta (float) visco-plastic coefficient (optional, default=0.0)

See also:
Notes

CastFuse Material

uniaxialMaterial(’Cast’, matTag, n, bo, h, fy, E, L, b, Ro, cR1, cR2, a1=s2*Pp/Kp, a2=1.0,

a3=a4*Pp/Kp, a4=1.0)
This command is used to construct a parallel material object made up of an arbitrary number of previously-
constructed UniaxialMaterial objects.

matTag integer tag identifying material

(int)
n (int) Number of yield fingers of the CSF-brace
bo Width of an individual yielding finger at its base of the CSF-brace
(float)
h (float) Thickness of an individual yielding finger
fy Yield strength of the steel material of the yielding finger
(float)
E (float) Modulus of elasticity of the steel material of the yielding finger
L (float) Height of an individual yielding finger
b (float) Strain hardening ratio
Ro Parameter that controls the Bauschinger effect. Recommended Values for $Ro=between 10 to
(float) 30
cR1 Parameter that controls the Bauschinger effect. Recommended Value cR1=0.925
(float)
cR2 Parameter that controls the Bauschinger effect. Recommended Value cR2=0.150
(float)
a1 isotropic hardening parameter, increase of compression yield envelope as proportion of yield
(float) strength after a plastic deformation of a2*(Pp/Kp)
a2 isotropic hardening parameter (see explanation under a1). (optional default = 1.0)
(float)
a3 isotropic hardening parameter, increase of tension yield envelope as proportion of yield strength
(float) after a plastic deformation of a4*(Pp/Kp)
a4 isotropic hardening parameter (see explanation under a3). (optional default = 1.0)
(float)

Gray et al. [1] showed that the monotonic backbone curve of a CSF-brace with known properties (n, bo, h, L, fy, E)
after yielding can be expressed as a close-form solution that is given by, 𝑃 = 𝑃𝑝 / cos(2𝑑/𝐿), in which 𝑑 is the axial
deformation of the brace at increment 𝑖 and 𝑃𝑝 is the yield strength of the CSF-brace and is given by the following
expression
𝑃𝑝 = 𝑛𝑏𝑜 ℎ2 𝑓𝑦 /4𝐿
The elastic stiffness of the CSF-brace is given by,

1.4. Model Commands 151

OpenSeesPy Documentation, Release 3.5.1.3

𝐾𝑝 = 𝑛𝑏𝑜 𝐸ℎ3 𝑓𝑦 /6𝐿3

See also:
Notes

ViscousDamper Material

uniaxialMaterial(’ViscousDamper’, matTag, K_el, Cd, alpha, LGap=0.0, NM=1, RelTol=1e-6,

AbsTol=1e-10, MaxHalf=15)
This command is used to construct a ViscousDamper material, which represents the Maxwell Model (linear
spring and nonlinear dashpot in series). The ViscousDamper material simulates the hysteretic response of non-
linear viscous dampers. An adaptive iterative algorithm has been implemented and validated to solve numeri-
cally the constitutive equations within a nonlinear viscous damper with a high-precision accuracy.

matTag integer tag identifying material

(int)
K_el Elastic stiffness of linear spring to model the axial flexibility of a viscous damper (e.g. combined
(float) stiffness of the supporting brace and internal damper portion)
Cd Damping coefficient
(float)
alpha Velocity exponent
(float)
LGap Gap length to simulate the gap length due to the pin tolerance
(float)
NM (int) Employed adaptive numerical algorithm (default value NM = 1; * 1 = Dormand-Prince54, * 2 =
6th order Adams-Bashforth-Moulton, * 3 = modified Rosenbrock Triple)
RelTol Tolerance for absolute relative error control of the adaptive iterative algorithm (default value
(float) 10^-6)
AbsTol Tolerance for absolute error control of adaptive iterative algorithm (default value 10^-10)
(float)
MaxHalf Maximum number of sub-step iterations within an integration step (default value 15)
(int)

See also:
Notes

BilinearOilDamper Material

uniaxialMaterial(’BilinearOilDamper’, matTag, K_el, Cd, Fr=1.0, p=1.0, LGap=0.0, NM=1,

RelTol=1e-6, AbsTol=1e-10, MaxHalf=15)
This command is used to construct a BilinearOilDamper material, which simulates the hysteretic response of
bilinear oil dampers with relief valve. Two adaptive iterative algorithms have been implemented and validated
to solve numerically the constitutive equations within a bilinear oil damper with a high-precision accuracy.

152 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

K_el (float) Elastic stiffness of linear spring to model the axial
flexibility of a viscous damper (e.g. combined stiff-
ness of the supporting brace and internal damper por-
tion)
Cd (float) Damping coefficient
Fr (float) Damper relief load (default=1.0, Damper property)
p (float) Post-relief viscous damping coefficient ratio (de-
fault=1.0, linear oil damper)
LGap (float) Gap length to simulate the gap length due to the pin
tolerance
NM (int) Employed adaptive numerical algorithm (default
value NM = 1;
• 1 = Dormand-Prince54,
• 2 = 6th order Adams-Bashforth-Moulton,
• 3 = modified Rosenbrock Triple)

RelTol (float) Tolerance for absolute relative error control of the

adaptive iterative algorithm (default value 10^-6)
AbsTol (float) Tolerance for absolute error control of adaptive iter-
ative algorithm (default value 10^-10)
MaxHalf (int) Maximum number of sub-step iterations within an
integration step (default value 15)

See also:
Notes

Modified Ibarra-Medina-Krawinkler Deterioration Model with Bilinear Hysteretic Response (Bilin Ma-
terial)

uniaxialMaterial(’Bilin’, matTag, K0, as_Plus, as_Neg, My_Plus, My_Neg, Lamda_S, Lamda_C,

Lamda_A, Lamda_K, c_S, c_C, c_A, c_K, theta_p_Plus, theta_p_Neg, theta_pc_Plus,
theta_pc_Neg, Res_Pos, Res_Neg, theta_u_Plus, theta_u_Neg, D_Plus, D_Neg, nFac-
tor=0.0)
This command is used to construct a bilin material. The bilin material simulates the modified Ibarra-Krawinkler
deterioration model with bilinear hysteretic response. Note that the hysteretic response of this material has been
calibrated with respect to more than 350 experimental data of steel beam-to-column connections and multivariate
regression formulas are provided to estimate the deterioration parameters of the model for different connection
types. These relationships were developed by Lignos and Krawinkler (2009, 2011) and have been adopted by
PEER/ATC (2010). The input parameters for this component model can be computed interactively from this
link. Use the module Component Model.

1.4. Model Commands 153

OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
K0 (float) elastic stiffness
as_Plus strain hardening ratio for positive loading direction
(float)
as_Neg strain hardening ratio for negative loading direction
(float)
My_Plus effective yield strength for positive loading direction
(float)
My_Neg effective yield strength for negative loading direction (negative value)
(float)
Lamda_S Cyclic deterioration parameter for strength deterioration [E_t=Lamda_S*M_y; set Lamda_S
(float) = 0 to disable this mode of deterioration]
Lamda_C Cyclic deterioration parameter for post-capping strength deterioration [E_t=Lamda_C*M_y;
(float) set Lamda_C = 0 to disable this mode of deterioration]
Lamda_A Cyclic deterioration parameter for acceleration reloading stiffness deterioration (is not a de-
(float) terioration mode for a component with Bilinear hysteretic response) [Input value is required,
but not used; set Lamda_A = 0].
Lamda_K Cyclic deterioration parameter for unloading stiffness deterioration [E_t=Lamda_K*M_y; set
(float) Lamda_k = 0 to disable this mode of deterioration]
c_S (float) rate of strength deterioration. The default value is 1.0.
c_C (float) rate of post-capping strength deterioration. The default value is 1.0.
c_A (float) rate of accelerated reloading deterioration. The default value is 1.0.
c_K (float) rate of unloading stiffness deterioration. The default value is 1.0.
theta_p_Pluspre-capping rotation for positive loading direction (often noted as plastic rotation capacity)
(float)
theta_p_Neg pre-capping rotation for negative loading direction (often noted as plastic rotation capacity)
(float) (positive value)
post-capping rotation for positive loading direction
theta_pc_Plus
(float)
theta_pc_Negpost-capping rotation for negative loading direction (positive value)
(float)
Res_Pos residual strength ratio for positive loading direction
(float)
Res_Neg residual strength ratio for negative loading direction (positive value)
(float)
theta_u_Plusultimate rotation capacity for positive loading direction
(float)
theta_u_Neg ultimate rotation capacity for negative loading direction (positive value)
(float)
D_Plus rate of cyclic deterioration in the positive loading direction (this parameter is used to create
(float) assymetric hysteretic behavior for the case of a composite beam). For symmetric hysteretic
response use 1.0.
D_Neg rate of cyclic deterioration in the negative loading direction (this parameter is used to create
(float) assymetric hysteretic behavior for the case of a composite beam). For symmetric hysteretic
response use 1.0.
nFactor elastic stiffness amplification factor, mainly for use with concentrated plastic hinge elements
(float) (optional, default = 0).

See also:
Notes

154 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Modified Ibarra-Medina-Krawinkler Deterioration Model with Peak-Oriented Hysteretic Response

(ModIMKPeakOriented Material)

uniaxialMaterial(’ModIMKPeakOriented’, matTag, K0, as_Plus, as_Neg, My_Plus, My_Neg, Lamda_S,

Lamda_C, Lamda_A, Lamda_K, c_S, c_C, c_A, c_K, theta_p_Plus, theta_p_Neg,
theta_pc_Plus, theta_pc_Neg, Res_Pos, Res_Neg, theta_u_Plus, theta_u_Neg,
D_Plus, D_Neg)
This command is used to construct a ModIMKPeakOriented material. This material simulates the modified
Ibarra-Medina-Krawinkler deterioration model with peak-oriented hysteretic response. Note that the hysteretic
response of this material has been calibrated with respect to 200 experimental data of RC beams in order to
estimate the deterioration parameters of the model. This information was developed by Lignos and Krawinkler
(2012). NOTE: before you use this material make sure that you have downloaded the latest OpenSees version.
A youtube video presents a summary of this model including the way to be used within openSees youtube link.

1.4. Model Commands 155

OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
K0 (float) elastic stiffness
as_Plus strain hardening ratio for positive loading direction
(float)
as_Neg strain hardening ratio for negative loading direction
(float)
My_Plus effective yield strength for positive loading direction
(float)
My_Neg effective yield strength for negative loading direction (negative value)
(float)
Lamda_S Cyclic deterioration parameter for strength deterioration [E_t=Lamda_S*M_y, see Lignos
(float) and Krawinkler (2011); set Lamda_S = 0 to disable this mode of deterioration]
Lamda_C Cyclic deterioration parameter for post-capping strength deterioration [E_t=Lamda_C*M_y,
(float) see Lignos and Krawinkler (2011); set Lamda_C = 0 to disable this mode of deterioration]
Lamda_A Cyclic deterioration parameter for accelerated reloading stiffness deterioration
(float) [E_t=Lamda_A*M_y, see Lignos and Krawinkler (2011); set Lamda_A = 0 to disable
this mode of deterioration]
Lamda_K Cyclic deterioration parameter for unloading stiffness deterioration [E_t=Lamda_K*M_y, see
(float) Lignos and Krawinkler (2011); set Lamda_K = 0 to disable this mode of deterioration]
c_S (float) rate of strength deterioration. The default value is 1.0.
c_C (float) rate of post-capping strength deterioration. The default value is 1.0.
c_A (float) rate of accelerated reloading deterioration. The default value is 1.0.
c_K (float) rate of unloading stiffness deterioration. The default value is 1.0.
theta_p_Pluspre-capping rotation for positive loading direction (often noted as plastic rotation capacity)
(float)
theta_p_Negpre-capping rotation for negative loading direction (often noted as plastic rotation capacity)
(float) (must be defined as a positive value)
post-capping rotation for positive loading direction
theta_pc_Plus
(float)
theta_pc_Negpost-capping rotation for negative loading direction (must be defined as a positive value)
(float)
Res_Pos residual strength ratio for positive loading direction
(float)
Res_Neg residual strength ratio for negative loading direction (must be defined as a positive value)
(float)
theta_u_Plusultimate rotation capacity for positive loading direction
(float)
theta_u_Negultimate rotation capacity for negative loading direction (must be defined as a positive value)
(float)
D_Plus rate of cyclic deterioration in the positive loading direction (this parameter is used to create
(float) assymetric hysteretic behavior for the case of a composite beam). For symmetric hysteretic
response use 1.0.
D_Neg rate of cyclic deterioration in the negative loading direction (this parameter is used to create
(float) assymetric hysteretic behavior for the case of a composite beam). For symmetric hysteretic
response use 1.0.

See also:
Notes

156 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Modified Ibarra-Medina-Krawinkler Deterioration Model with Pinched Hysteretic Response (Mod-

IMKPinching Material)

uniaxialMaterial(’ModIMKPinching’, matTag, K0, as_Plus, as_Neg, My_Plus, My_Neg, FprPos,

FprNeg, A_pinch, Lamda_S, Lamda_C, Lamda_A, Lamda_K, c_S, c_C, c_A,
c_K, theta_p_Plus, theta_p_Neg, theta_pc_Plus, theta_pc_Neg, Res_Pos, Res_Neg,
theta_u_Plus, theta_u_Neg, D_Plus, D_Neg)
This command is used to construct a ModIMKPinching material. This material simulates the modified Ibarra-
Medina-Krawinkler deterioration model with pinching hysteretic response. NOTE: before you use this mate-
rial make sure that you have downloaded the latest OpenSees version. A youtube video presents a summary
of this model including the way to be used within openSees youtube link.

1.4. Model Commands 157

OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
K0 (float) elastic stiffness
as_Plus strain hardening ratio for positive loading direction
(float)
as_Neg strain hardening ratio for negative loading direction
(float)
My_Plus effective yield strength for positive loading direction
(float)
My_Neg effective yield strength for negative loading direction (Must be defined as a negative value)
(float)
FprPos Ratio of the force at which reloading begins to force corresponding to the maximum historic
(float) deformation demand (positive loading direction)
FprNeg Ratio of the force at which reloading begins to force corresponding to the absolute maximum
(float) historic deformation demand (negative loading direction)
A_pinch Ratio of reloading stiffness
(float)
Lamda_S Cyclic deterioration parameter for strength deterioration [E_t=Lamda_S*M_y, see Lignos
(float) and Krawinkler (2011); set Lamda_S = 0 to disable this mode of deterioration]
Lamda_C Cyclic deterioration parameter for post-capping strength deterioration [E_t=Lamda_C*M_y,
(float) see Lignos and Krawinkler (2011); set Lamda_C = 0 to disable this mode of deterioration]
Lamda_A Cyclic deterioration parameter for accelerated reloading stiffness deterioration
(float) [E_t=Lamda_A*M_y, see Lignos and Krawinkler (2011); set Lamda_A = 0 to disable
this mode of deterioration]
Lamda_K Cyclic deterioration parameter for unloading stiffness deterioration [E_t=Lamda_K*M_y, see
(float) Lignos and Krawinkler (2011); set Lamda_K = 0 to disable this mode of deterioration]
c_S (float) rate of strength deterioration. The default value is 1.0.
c_C (float) rate of post-capping strength deterioration. The default value is 1.0.
c_A (float) rate of accelerated reloading deterioration. The default value is 1.0.
c_K (float) rate of unloading stiffness deterioration. The default value is 1.0.
theta_p_Pluspre-capping rotation for positive loading direction (often noted as plastic rotation capacity)
(float)
theta_p_Negpre-capping rotation for negative loading direction (often noted as plastic rotation capacity)
(float) (must be defined as a positive value)
post-capping rotation for positive loading direction
theta_pc_Plus
(float)
theta_pc_Negpost-capping rotation for negative loading direction (must be defined as a positive value)
(float)
Res_Pos residual strength ratio for positive loading direction
(float)
Res_Neg residual strength ratio for negative loading direction (must be defined as a positive value)
(float)
theta_u_Plusultimate rotation capacity for positive loading direction
(float)
theta_u_Negultimate rotation capacity for negative loading direction (must be defined as a positive value)
(float)
D_Plus rate of cyclic deterioration in the positive loading direction (this parameter is used to create
(float) assymetric hysteretic behavior for the case of a composite beam). For symmetric hysteretic
response use 1.0.
D_Neg rate of cyclic deterioration in the negative loading direction (this parameter is used to create
(float) assymetric hysteretic behavior for the case of a composite beam). For symmetric hysteretic
response use 1.0.

158 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

See also:
Notes

SAWS Material

uniaxialMaterial(’SAWS’, matTag, F0, FI, DU, S0, R1, R2, R3, R4, alpha, beta)
This file contains the class definition for SAWSMaterial. SAWSMaterial provides the implementation of a
one-dimensional hysteretic model develeped as part of the CUREe Caltech wood frame project.

matTag integer tag identifying material

(int)
F0 (float) Intercept strength of the shear wall spring element for the asymtotic line to the envelope curve
F0 > FI > 0
FI (float) Intercept strength of the spring element for the pinching branch of the hysteretic curve. (FI >
0).
DU (float) Spring element displacement at ultimate load. (DU > 0).
S0 (float) Initial stiffness of the shear wall spring element (S0 > 0).
R1 (float) Stiffness ratio of the asymptotic line to the spring element envelope curve. The slope of this
line is R1 S0. (0 < R1 < 1.0).
R2 (float) Stiffness ratio of the descending branch of the spring element envelope curve. The slope of
this line is R2 S0. ( R2 < 0).
R3 (float) Stiffness ratio of the unloading branch off the spring element envelope curve. The slope of this
line is R3 S0. ( R3 1).
R4 (float) Stiffness ratio of the pinching branch for the spring element. The slope of this line is R4 S0. (
R4 > 0).
alpha Stiffness degradation parameter for the shear wall spring element. (ALPHA > 0).
(float)
beta Stiffness degradation parameter for the spring element. (BETA > 0).
(float)

See also:
Notes

BarSlip Material

uniaxialMaterial(’BarSlip’, matTag, fc, fy, Es, fu, Eh, db, ld, nb, depth, height, ancLratio=1.0, bsFlag,
type, damage=’Damage’, unit=’psi’)
This command is used to construct a uniaxial material that simulates the bar force versus slip response of a
reinforcing bar anchored in a beam-column joint. The model exhibits degradation under cyclic loading. Cyclic
degradation of strength and stiffness occurs in three ways: unloading stiffness degradation, reloading stiffness
degradation, strength degradation.

1.4. Model Commands 159

OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
fc (float) positive floating point value defining the compressive strength of the concrete in which the
reinforcing bar is anchored
fy (float) positive floating point value defining the yield strength of the reinforcing steel
Es (float) floating point value defining the modulus of elasticity of the reinforcing steel
fu (float) positive floating point value defining the ultimate strength of the reinforcing steel
Eh (float) floating point value defining the hardening modulus of the reinforcing steel
ld (float) floating point value defining the development length of the reinforcing steel
db (float) point value defining the diameter of reinforcing steel
nb (int) an integer defining the number of anchored bars
depth floating point value defining the dimension of the member (beam or column) perpendicular to
(float) the dimension of the plane of the paper
height floating point value defining the height of the flexural member, perpendicular to direction in
(float) which the reinforcing steel is placed, but in the plane of the paper
ancLratio floating point value defining the ratio of anchorage length used for the reinforcing bar to the
(float) dimension of the joint in the direction of the reinforcing bar (optional, default: 1.0)
bsFlag string indicating relative bond strength for the anchored reinforcing bar (options: 'Strong'
(str) or 'Weak')
type (str) string indicating where the reinforcing bar is placed. (options: 'beamtop', 'beambot' or
'column')
damage string indicating type of damage:whether there is full damage in the material or no damage
(str) (optional, options: 'Damage', 'NoDamage' ; default: 'Damage')
unit (str) string indicating the type of unit system used (optional, options: 'psi', 'MPa', 'Pa',
'psf', 'ksi', 'ksf') (default: 'psi' / 'MPa')

See also:
Notes

Bond SP01 - - Strain Penetration Model for Fully Anchored Steel Reinforcing Bars

uniaxialMaterial(’Bond_SP01’, matTag, Fy, Sy, Fu, Su, b, R)

This command is used to construct a uniaxial material object for capturing strain penetration effects at the
column-to-footing, column-to-bridge bent caps, and wall-to-footing intersections. In these cases, the bond slip
associated with strain penetration typically occurs along a portion of the anchorage length. This model can also
be applied to the beam end regions, where the strain penetration may include slippage of the bar along the entire
anchorage length, but the model parameters should be chosen appropriately.
This model is for fully anchored steel reinforcement bars that experience bond slip along a portion of the an-
chorage length due to strain penetration effects, which are usually the case for column and wall longitudinal
bars anchored into footings or bridge joints

matTag (int) integer tag identifying material

Fy (float) Yield strength of the reinforcement steel
Sy (float) Rebar slip at member interface under yield stress. (see NOTES below)
Fu (float) Ultimate strength of the reinforcement steel
Su (float) Rebar slip at the loaded end at the bar fracture strength
b (float) Initial hardening ratio in the monotonic slip vs. bar stress response (0.3~0.5)
R (float) Pinching factor for the cyclic slip vs. bar response (0.5~1.0)

See also:

160 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Notes

Fatigue Material

uniaxialMaterial(’Fatigue’, matTag, otherTag, ’-E0’, E0=0.191, ’-m’, m=-0.458, ’-min’, min=-1e16,

’-max’, max=1e16)
The fatigue material uses a modified rainflow cycle counting algorithm to accumulate damage in a material
using Miner’s Rule. Element stress/strain relationships become zero when fatigue life is exhausted.

matTag (int) integer tag identifying material

otherTag (float) Unique material object integer tag for the material that is being wrapped
E0 (float) Value of strain at which one cycle will cause failure (default 0.191)
m (float) Slope of Coffin-Manson curve in log-log space (default -0.458)
min (float) Global minimum value for strain or deformation (default -1e16)
max (float) Global maximum value for strain or deformation (default 1e16)

See also:
Notes

Impact Material

uniaxialMaterial(’ImpactMaterial’, matTag, K1, K2, sigy, gap)

This command is used to construct an impact material object

matTag (int) integer tag identifying material

K1 (float) initial stiffness
K2 (float) secondary stiffness
sigy (float) yield displacement
gap (float) initial gap

See also:
Notes

Hyperbolic Gap Material

uniaxialMaterial(’HyperbolicGapMaterial’, matTag, Kmax, Kur, Rf, Fult, gap)

This command is used to construct a hyperbolic gap material object.

matTag (int) integer tag identifying material

Kmax (float) initial stiffness
Kur (float) unloading/reloading stiffness
Rf (float) failure ratio
Fult (float) ultimate (maximum) passive resistance
gap (float) initial gap

Note:

1.4. Model Commands 161

OpenSeesPy Documentation, Release 3.5.1.3

1. This material is implemented as a compression-only gap material. Fult and gap should be input as negative
values.
2. Recomended Values:
• Kmax = 20300 kN/m of abutment width
• Kcur = Kmax
• Rf = 0.7
• Fult = -326 kN per meter of abutment width
• gap = -2.54 cm

See also:
Notes

Limit State Material

uniaxialMaterial(’LimitState’, matTag, s1p, e1p, s2p, e2p, s3p, e3p, s1n, e1n, s2n, e2n, s3n, e3n, pinchX,
pinchY, damage1, damage2, beta, curveTag, curveType)
This command is used to construct a uniaxial hysteretic material object with pinching of force and deformation,
damage due to ductility and energy, and degraded unloading stiffness based on ductility. Failure of the material
is defined by the associated Limit Curve.

matTag (int) integer tag identifying material

s1p e1p stress and strain (or force & deformation) at first point of the envelope in the positive
(float) direction
s2p e2p stress and strain (or force & deformation) at second point of the envelope in the positive
(float) direction
s3p e3p stress and strain (or force & deformation) at third point of the envelope in the positive
(float) direction
s1n e1n stress and strain (or force & deformation) at first point of the envelope in the negative
(float) direction
s2n e2n stress and strain (or force & deformation) at second point of the envelope in the negative
(float) direction
s3n e3n stress and strain (or force & deformation) at third point of the envelope in the negative
(float) direction
pinchX pinching factor for strain (or deformation) during reloading
(float)
pinchY pinching factor for stress (or force) during reloading
(float)
damage1 damage due to ductility: D1(m-1)
(float)
damage2 damage due to energy: D2(Ei/Eult)
(float)
beta (float) power used to determine the degraded unloading stiffness based on ductility, m-b (optional,
default=0.0)
curveTag an integer tag for the Limit Curve defining the limit surface
(int)
curveType an integer defining the type of LimitCurve (0 = no curve, 1 = axial curve, all other curves
(int) can be any other integer)

162 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Note:
• negative backbone points should be entered as negative numeric values

See also:
Notes

MinMax Material

uniaxialMaterial(’MinMax’, matTag, otherTag, ’-min’, minStrain=1e-16, ’-max’, maxStrain=1e16)

This command is used to construct a MinMax material object. This stress-strain behaviour for this material is
provided by another material. If however the strain ever falls below or above certain threshold values, the other
material is assumed to have failed. From that point on, values of 0.0 are returned for the tangent and stress.

matTag (int) integer tag identifying material

otherTag (float) tag of the other material
minStrain (float) minimum value of strain. optional default = -1.0e16.
maxStrain (float) max value of strain. optional default = 1.0e16.

See also:
Notes

ElasticBilin Material

uniaxialMaterial(’ElasticBilin’, matTag, EP1, EP2, epsP2, EN1=EP1, EN2=EP2, epsN2=-epsP2)

This command is used to construct an elastic bilinear uniaxial material object. Unlike all other bilinear materials,
the unloading curve follows the loading curve exactly.

matTag (int) integer tag identifying material

EP1 (float) tangent in tension for stains: 0 <= strains <= epsP2
EP2 (float) tangent when material in tension with strains > epsP2
epsP2 (float) strain at which material changes tangent in tension.
EN1 (float) optional, default = EP1. tangent in compression for stains: 0 < strains <= epsN2
EN2 (float) optional, default = EP2. tangent in compression with strains < epsN2
epsN2 (float) optional, default = -epsP2. strain at which material changes tangent in compression.

Note: eps0 can not be controlled. It is always zero.

See also:
Notes

ElasticMultiLinear Material

uniaxialMaterial(’ElasticMultiLinear’, matTag, eta=0.0, ’-strain’, *strain, ’-stress’, *stress)

This command is used to construct a multi-linear elastic uniaxial material object. The nonlinear stress-strain
relationship is given by a multi-linear curve that is define by a set of points. The behavior is nonlinear but it is

1.4. Model Commands 163

OpenSeesPy Documentation, Release 3.5.1.3

elastic. This means that the material loads and unloads along the same curve, and no energy is dissipated. The
slope given by the last two specified points on the positive strain axis is extrapolated to infinite positive strain.
Similarly, the slope given by the last two specified points on the negative strain axis is extrapolated to infinite
negative strain. The number of provided strain points needs to be equal to the number of provided stress points.

matTag (int) integer tag identifying material

eta (float) damping tangent (optional, default=0.0)
strain (list (float)) list of strain points along stress-strain curve
stress (list (float)) list of stress points along stress-strain curve

See also:
Notes

MultiLinear

uniaxialMaterial(’MultiLinear’, matTag, *pts)

This command is used to construct a uniaxial multilinear material object.

matTag (int) integer tag identifying material

pts (list a list of strain and stress points
(float)) pts = [strain1, stress1, strain2, stress2, ..., ]

See also:
Notes

Initial Strain Material

uniaxialMaterial(’InitStrainMaterial’, matTag, otherTag, initStrain)

This command is used to construct an Initial Strain material object. The stress-strain behaviour for this material
is defined by another material. Initial Strain Material enables definition of initial strains for the material under
consideration. The stress that corresponds to the initial strain will be calculated from the other material.

matTag (int) integer tag identifying material

otherTag (int) tag of the other material
initStrain (float) initial strain

See also:
Notes

Initial Stress Material

uniaxialMaterial(’InitStressMaterial’, matTag, otherTag, initStress)

This command is used to construct an Initial Stress material object. The stress-strain behaviour for this material
is defined by another material. Initial Stress Material enables definition of initial stress for the material under
consideration. The strain that corresponds to the initial stress will be calculated from the other material.

164 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

otherTag (float) tag of the other material
initStress (float) initial stress

See also:
Notes

PathIndependent Material

uniaxialMaterial(’PathIndependent’, matTag, OtherTag)

This command is to create a PathIndependent material

matTag (int) integer tag identifying material

OtherTag (int) a pre-defined material

Pinching4 Material

uniaxialMaterial(’Pinching4’, matTag, ePf1, ePd1, ePf2, ePd2, ePf3, ePd3, ePf4, ePd4, <eNf1, eNd1,
eNf2, eNd2, eNf3, eNd3, eNf4, eNd4>, rDispP, rForceP, uForceP, <rDispN, rForceN,
uForceN>, gK1, gK2, gK3, gK4, gKLim, gD1, gD2, gD3, gD4, gDLim, gF1, gF2, gF3,
gF4, gFLim, gE, dmgType)
This command is used to construct a uniaxial material that represents a ‘pinched’ load-deformation response
and exhibits degradation under cyclic loading. Cyclic degradation of strength and stiffness occurs in three ways:
unloading stiffness degradation, reloading stiffness degradation, strength degradation.

1.4. Model Commands 165

OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

ePf1 ePf2 floating point values defining force points on the positive response envelope
ePf3 ePf4
(float)
ePd1 ePd2 floating point values defining deformation points on the positive response envelope
ePd3 ePd4
(float)
eNf1 eNf2 floating point values defining force points on the negative response envelope
eNf3 eNf4
(float)
eNd1 eNd2 floating point values defining deformation points on the negative response envelope
eNd3 eNd4
(float)
rDispP floating point value defining the ratio of the deformation at which reloading occurs to the
(float) maximum historic deformation demand
fFoceP floating point value defining the ratio of the force at which reloading begins to force cor-
(float) responding to the maximum historic deformation demand
uForceP floating point value defining the ratio of strength developed upon unloading from negative
(float) load to the maximum strength developed under monotonic loading
rDispN floating point value defining the ratio of the deformation at which reloading occurs to the
(float) minimum historic deformation demand
fFoceN floating point value defining the ratio of the force at which reloading begins to force cor-
(float) responding to the minimum historic deformation demand
uForceN floating point value defining the ratio of strength developed upon unloading from negative
(float) load to the minimum strength developed under monotonic loading
gK1 gK2 gK3 floating point values controlling cyclic degradation model for unloading stiffness degrada-
gK4 gKLim tion
(float)
gD1 gD2 gD3 floating point values controlling cyclic degradation model for reloading stiffness degrada-
gD4 gDLim tion
(float)
gF1 gF2 gF3 floating point values controlling cyclic degradation model for strength degradation
gF4 gFLim
(float)
gE (float) floating point value used to define maximum energy dissipation under cyclic loading. Total
energy dissipation capacity is defined as this factor multiplied by the energy dissipated
under monotonic loading.
dmgType (str) string to indicate type of damage (option: 'cycle', 'energy')

See also:
Notes

Engineered Cementitious Composites Material

uniaxialMaterial(’ECC01’, matTag, sigt0, epst0, sigt1, epst1, epst2, sigc0, epsc0, epsc1, alphaT1, al-
phaT2, alphaC, alphaCU, betaT, betaC)
This command is used to construct a uniaxial Engineered Cementitious Composites (ECC)material object based
on the ECC material model of Han, et al. (see references). Reloading in tension and compression is linear.

166 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

sigt0 (float) tensile cracking stress
epst0 (float) strain at tensile cracking stress
sigt1 (float) peak tensile stress
epst1 (float) strain at peak tensile stress
epst2 (float) ultimate tensile strain
sigc0 (float) compressive strength (see NOTES)
epsc0 (float) strain at compressive strength (see NOTES)
epsc1 (float) ultimate compressive strain (see NOTES)
alphaT1 (float) exponent of the unloading curve in tensile strain hardening region
alphaT2 (float) exponent of the unloading curve in tensile softening region
alphaC (float) exponent of the unloading curve in the compressive softening
alphaCU (float) exponent of the compressive softening curve (use 1 for linear softening)
betaT (float) parameter to determine permanent strain in tension
betaC (float) parameter to determine permanent strain in compression

See also:
Notes

SelfCentering Material

uniaxialMaterial(’SelfCentering’, matTag, k1, k2, sigAct, beta, epsSlip=0, epsBear=0, rBear=k1)

This command is used to construct a uniaxial self-centering (flag-shaped) material object with optional non-
recoverable slip behaviour and an optional stiffness increase at high strains (bearing behaviour).

matTag (int) integer tag identifying material

k1 (float) Initial Stiffness
k2 (float) Post-Activation Stiffness (0< k2``< ``k1)
sigAct (float) Forward Activation Stress/Force
beta (float) Ratio of Forward to Reverse Activation Stress/Force
epsSlip (float) slip Strain/Deformation (if epsSlip = 0, there will be no slippage)
epsBear (float) Bearing Strain/Deformation (if epsBear = 0, there will be no bearing)
rBear (float) Ratio of Bearing Stiffness to Initial Stiffness k1

See also:
Notes

Viscous Material

uniaxialMaterial(’Viscous’, matTag, C, alpha)

This command is used to construct a uniaxial viscous material object. stress =C(strain-rate)^alpha

matTag (int) integer tag identifying material

C (float) damping coeficient
alpha (float) power factor (=1 means linear damping)

Note:

1.4. Model Commands 167

OpenSeesPy Documentation, Release 3.5.1.3

1. This material can only be assigned to truss and zeroLength elements.

2. This material can not be combined in parallel/series with other materials. When defined in parallel with other
materials it is ignored.

See also:
Notes

BoucWen Material

uniaxialMaterial(’BoucWen’, matTag, alpha, ko, n, gamma, beta, Ao, deltaA, deltaNu, deltaEta)
This command is used to construct a uniaxial Bouc-Wen smooth hysteretic material object. This material model
is an extension of the original Bouc-Wen model that includes stiffness and strength degradation (Baber and
Noori (1985)).

matTag (int) integer tag identifying material

alpha (float) ratio of post-yield stiffness to the initial elastic stiffenss (0< alpha <1)
ko (float) initial elastic stiffness
n (float) parameter that controls transition from linear to nonlinear range (as n increases the transi-
tion becomes sharper; n is usually grater or equal to 1)
gamma beta parameters that control shape of hysteresis loop; depending on the values of gamma and
(float) beta softening, hardening or quasi-linearity can be simulated (look at the NOTES)
Ao deltaA parameters that control tangent stiffness
(float)
deltaNu parameters that control material degradation
deltaEta
(float)

See also:
Notes

BWBN Material

uniaxialMaterial(’BWBN’, matTag, alpha, ko, n, gamma, beta, Ao, q, zetas, p, Shi, deltaShi, lambda,
tol, maxIter)
This command is used to construct a uniaxial Bouc-Wen pinching hysteretic material object. This material
model is an extension of the original Bouc-Wen model that includes pinching (Baber and Noori (1986) and
Foliente (1995)).

168 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

alpha (float) ratio of post-yield stiffness to the initial elastic stiffenss (0< alpha <1)
ko (float) initial elastic stiffness
n (float) parameter that controls transition from linear to nonlinear range (as n increases the
transition becomes sharper; n is usually grater or equal to 1)
gamma beta (float) parameters that control shape of hysteresis loop; depending on the values of gamma
and beta softening, hardening or quasi-linearity can be simulated (look at the
BoucWen Material)
Ao (float) parameter that controls tangent stiffness
q zetas p parameters that control pinching
Shi deltaShi
lambda (float)
tol (float) tolerance
maxIter (float) maximum iterations

See also:
Notes

KikuchiAikenHDR Material

uniaxialMaterial(’KikuchiAikenHDR’, matTag, tp, ar, hr, <’-coGHU’, cg, ch, cu>, <’-coMSS’, rs, rf>)
This command is used to construct a uniaxial KikuchiAikenHDR material object. This material model produces
nonlinear hysteretic curves of high damping rubber bearings (HDRs).

matTag integer tag identifying material

(int)
tp (str) rubber type (see note 1)
ar (float) area of rubber [unit: m^2] (see note 2)
hr (float) total thickness of rubber [unit: m] (see note 2)
cg ch cu correction coefficients for equivalent shear modulus (cg), equivalent viscous daming ratio
(float) (ch), ratio of shear force at zero displacement (cu).
rs rf reduction rate for stiffness (rs) and force (rf) (see note 3)
(float)

Note:
1) Following rubber types for tp are available:
• 'X0.6' Bridgestone X0.6, standard compressive stress, up to 400% shear strain
• 'X0.6-0MPa' Bridgestone X0.6, zero compressive stress, up to 400% shear strain
• 'X0.4' Bridgestone X0.4, standard compressive stress, up to 400% shear strain
• 'X0.4-0MPa' Bridgestone X0.4, zero compressive stress, up to 400% shear strain
• 'X0.3' Bridgestone X0.3, standard compressive stress, up to 400% shear strain
• 'X0.3-0MPa' Bridgestone X0.3, zero compressive stress, up to 400% shear strain
2) This material uses SI unit in calculation formula. ar and hr must be converted into [m^2] and [m], respectively.

1.4. Model Commands 169

OpenSeesPy Documentation, Release 3.5.1.3

3) rs and rf areavailable if this material is applied to multipleShearSpring (MSS) element. Recommended values
1 1
are rs = ∑︀𝑛−1 sin(𝜋*𝑖/𝑛) 2
and rf = ∑︀𝑛−1 sin(𝜋*𝑖/𝑛) , where n is the number of springs in the MSS. For example,
𝑖=0 𝑖=0
when n=8, rs =0.2500, rf =0.1989.

See also:
Notes

KikuchiAikenLRB Material

uniaxialMaterial(’KikuchiAikenLRB’, matTag, type, ar, hr, gr, ap, tp, alph, beta, <’-T’, temp>, <’-
coKQ’, rk, rq>, <’-coMSS’, rs, rf>)
This command is used to construct a uniaxial KikuchiAikenLRB material object. This material model produces
nonlinear hysteretic curves of lead-rubber bearings.

matTag (int) integer tag identifying material

type (int) rubber type (see note 1)
ar (float) area of rubber [unit: m^2]
hr (float) total thickness of rubber [unit: m]
gr (float) shear modulus of rubber [unit: N/m^2]
ap (float) area of lead plug [unit: m^2]
tp (float) yield stress of lead plug [unit: N/m^2]
alph (float) shear modulus of lead plug [unit: N/m^2]
beta (float) ratio of initial stiffness to yielding stiffness
temp (float) temperature [unit: °C]
rk rq (float) reduction rate for yielding stiffness ( rk) and force at zero displacement ( rq)
rs rf (float) reduction rate for stiffness ( rs) and force ( rf) (see note 3)

Note:
1) Following rubber types for type are available:
• 1 lead-rubber bearing, up to 400% shear strain [Kikuchi et al., 2010 & 2012]
2) This material uses SI unit in calculation formula. Input arguments must be converted into [m], [m^2], [N/m^2].
3) rs and rf are available if this material is applied to multipleShearSpring (MSS) element. Recommended
1
values are rs = ∑︀𝑛−1 sin(𝜋*𝑖/𝑛) 2
and rf = ∑︀ 𝑛−11sin(𝜋*𝑖/𝑛) , where n is the number of springs in the MSS.
𝑖=0 𝑖=0
For example, when n=8, rs = 0.2500 and rf = 0.1989.

See also:
Notes

AxialSp Material

uniaxialMaterial(’AxialSp’, matTag, sce, fty, fcy, <bte, bty, bcy, fcr>)

This command is used to construct a uniaxial AxialSp material object. This material model produces axial
stress-strain curve of elastomeric bearings.

170 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

sce (float) compressive modulus
fty fcy (float) yield stress under tension ( fty) and compression ( fcy) (see note 1)
bte bty bcy reduction rate for tensile elastic range ( bte), tensile yielding ( bty) and compressive
(float) yielding ( bcy) (see note 1)
fcr (float) target point stress (see note 1)

Note:
1. Input parameters are required to satisfy followings.
fcy < 0.0 < fty
0.0 <= bty < bte <= 1.0
0.0 <= bcy <= 1.0
fcy <= fcr <= 0.0

See also:
Notes

AxialSpHD Material

uniaxialMaterial(’AxialSpHD’, matTag, sce, fty, fcy, <bte, bty, bth, bcy, fcr, ath>)
This command is used to construct a uniaxial AxialSpHD material object. This material model produces axial
stress-strain curve of elastomeric bearings including hardening behavior.

matTag (int) integer tag identifying material

sce (float) compressive modulus
fty fcy (float) yield stress under tension (fty) and compression (fcy) (see note 1)
bte bty bth reduction rate for tensile elastic range (bte), tensile yielding (bty), tensile hardening (
bcy (float) bth) and compressive yielding (bcy) (see note 1)
fcr (float) target point stress (see note 1)
ath (float) hardening strain ratio to yield strain

Note:
1. Input parameters are required to satisfy followings.
fcy < 0.0 < fty
0.0 <= bty < bth < bte <= 1.0
0.0 <= bcy <= 1.0
fcy <= fcr <= 0.0
1.0 <= ath

See also:
Notes

1.4. Model Commands 171

OpenSeesPy Documentation, Release 3.5.1.3

Pinching Limit State Material

This command is used to construct a uniaxial material that simulates a pinched load-deformation response and exhibits
degradation under cyclic loading. This material works with the RotationShearCurve limit surface that can monitor a
key deformation and/or a key force in an associated frame element and trigger a degrading behavior in this material
when a limiting value of the deformation and/or force are reached. The material can be used in two modes: 1) direct in-
put mode, where pinching and damage parameters are directly input; and 2) calibrated mode for shear-critical concrete
columns, where only key column properties are input for model to fully define pinching and damage parameters.
uniaxialMaterial(’PinchingLimitStateMaterial’, matTag, nodeT, nodeB, driftAxis, Kelas, crvTyp, crv-
Tag, YpinchUPN, YpinchRPN, XpinchRPN, YpinchUNP, YpinchRNP, XpinchRNP,
dmgStrsLimE, dmgDispMax, dmgE1, dmgE2, dmgE3, dmgE4, dmgELim, dmgR1,
dmgR2, dmgR3, dmgR4, dmgRLim, dmgRCyc, dmgS1, dmgS2, dmgS3, dmgS4,
dmgSLim, dmgSCyc)
MODE 1: Direct Input

172 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
nodeTinteger node tag to define the first node at the extreme end of the associated flexural frame member
(int) (L3 or D5 in Figure)
nodeBinteger node tag to define the last node at the extreme end of the associated flexural frame member
(int) (L2 or D2 in Figure)
integer to indicate the drift axis in which lateral-strength degradation will occur. This axis should be
driftAxis
(int) orthogonal to the axis of measured rotation (see rotAxis in Rotation Shear Curve definition)
driftAxis = 1 - Drift along the x-axis driftAxis = 2 - Drift along the y-axis driftAxis = 3
- Drift along the z-axis
Kelasfloating point value to define the initial material elastic stiffness (Kelastic); Kelas > 0
(float)
crvTyp integer flag to indicate the type of limit curve associated with this material.
(int) crvTyp = 0 - No limit curve crvTyp = 1 - axial limit curve crvTyp = 2 - RotationShearCurve
crvTag integer tag for the unique limit curve object associated with this material
(int)
floating point unloading force pinching factor for loading in the negative direction. Note: This value
YpinchUPN
(float) must be between zero and unity
floating point reloading force pinching factor for loading in the negative direction. Note: This value
YpinchRPN
(float) must be between negative one and unity
floating point reloading displacement pinching factor for loading in the negative direction. Note:
XpinchRPN
(float) This value must be between negative one and unity
floating point unloading force pinching factor for loading in the positive direction. Note: This value
YpinchUNP
(float) must be between zero and unity
floating point reloading force pinching factor for loading in the positive direction. Note: This value
YpinchRNP
(float) must be between negative one and unity
floating point reloading displacement pinching factor for loading in the positive direction. Note: This
XpinchRNP
(float) value must be between negative one and unity
floating point force limit for elastic stiffness damage (typically defined as the lowest of shear strength
dmgStrsLimE
(float) or shear at flexrual yielding). This value is used to compute the maximum deformation at flexural
yield (𝛿max Eq. 1) and using the initial elastic stiffness (Kelastic) the monotonic energy (Emono Eq.
1) to yield. Input 1 if this type of damage is not required and set dmgE1, dmgE2, dmgE3, dmgE4,
and dmgELim to zero
floating point for ultimate drift at failure (𝛿max Eq. 1) and is used for strength and stiffness damage.
dmgDispMax
(float) This value is used to compute the monotonic energy at axial failure (Emono Eq. 2) by computing the
area under the backbone in the positive loading direction up to 𝛿max. Input 1 if this type of damage is
not required and set dmgR1, dmgR2, dmgR3, dmgR4, and dmgRLim to zero for reloading stiffness
damage. Similarly set dmgS1, dmgS2, dmgS3, dmgS4, and dmgSLim to zero if reloading strength
damage is not required
dmgE1floating point elastic stiffness damage factors 𝛼1,𝛼2,𝛼3,𝛼4 shown in Eq. 1
dmgE2
(float)
dmgE3floating point elastic stiffness damage factors 𝛼1,𝛼2,𝛼3,𝛼4 shown in Eq. 1
dmgE4
(float)
dmgELim floating point elastic stiffness damage limit Dlim shown in Eq. 1; Note: This value must be between
(float) zero and unity
dmgR1floating point reloading stiffness damage factors 𝛼1,𝛼2,𝛼3,𝛼4 shown in Eq. 1
dmgR2
(float)
dmgR3floating point reloading stiffness damage factors 𝛼1,𝛼2,𝛼3,𝛼4 shown in Eq. 1
dmgR4
(float)
dmgRLim floating point reloading stiffness damage limit Dlim shown in Eq. 1; Note: This value must be
(float) between zero and unity
1.4. Model Commands 173
dmgRCyc floating point cyclic reloading stiffness damage index; Note: This value must be between zero and
(float) unity
dmgS1floating point backbone strength damage factors 𝛼1,𝛼2,𝛼3,𝛼4 shown in Eq. 1
OpenSeesPy Documentation, Release 3.5.1.3

uniaxialMaterial(’PinchingLimitStateMaterial’, matTag, dnodeT, nodeB, driftAxis, Kelas, crvTyp, crv-

Tag, eleTag, b, d, h, a, st, As, Acc, ld, db, rhot, fc, fy, fyt)
MODE 2: Calibrated Model for Shear-Critical Concrete Columns

174 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
nodeTinteger node tag to define the first node at the extreme end of the associated flexural frame member
(int) (L3 or D5 in Figure)
nodeBinteger node tag to define the last node at the extreme end of the associated flexural frame member
(int) (L2 or D2 in Figure)
integer to indicate the drift axis in which lateral-strength degradation will occur. This axis should be
driftAxis
(int) orthogonal to the axis of measured rotation (see rotAxis` in Rotation Shear Curve definition)
driftAxis = 1 - Drift along the x-axis driftAxis = 2 - Drift along the y-axis driftAxis = 3
- Drift along the z-axis
Kelasfloating point value to define the shear stiffness (Kelastic) of the shear spring prior to shear failure
(float) Kelas = -4 - Shear stiffness calculated assuming double curvature and shear springs at both column
element ends
Kelas = -3 - Shear stiffness calculated assuming double curvature and a shear spring at one column
element end
Kelas = -2 - Shear stiffness calculated assuming single curvature and shear springs at both column
element ends
Kelas = -1 - Shear stiffness calculated assuming single curvature and a shear spring at one column
element end
Kelas > 0 - Shear stiffness is the input value
Note: integer inputs allow the model to know whether column height equals the shear span (cantelever)
or twice the shear span (double curvature). For columns in frames, input the value for the case that
best approximates column end conditions or manually input shear stiffness (typically double curvature
better estimates framed column behavior)
crvTag integer tag for the unique limit curve object associated with this material
(int)
eleTag integer element tag to define the associated beam-column element used to extract axial load
(int)
b floating point column width (inches)
(float)
d floating point column depth (inches)
(float)
h floating point column height (inches)
(float)
a floating point shear span length (inches)
(float)
st floating point transverse reinforcement spacing (inches) along column height
(float)
As floating point total area (inches squared) of longitudinal steel bars in section
(float)
Acc floating point gross confined concrete area (inches squared) bounded by the transverse reinforcement
(float) in column section
ld floating point development length (inches) of longitudinal bars using ACI 318-11 Eq. 12-1 and Eq.
(float) 12-2
db floating point diameter (inches) of longitudinal bars in column section
(float)
rhot floating point transverse reinforcement ratio (Ast/st.db)
(float)
f'c floating point concrete compressive strength (ksi)
(float)
fy floating point longitudinal steel yield strength (ksi)
(float)
fyt floating point transverse steel yield strength (ksi)
(float)
1.4. Model Commands 175
OpenSeesPy Documentation, Release 3.5.1.3

See also:
Notes

CFSWSWP Wood-Sheathed Cold-Formed Steel Shear Wall Panel

uniaxialMaterial(’CFSWSWP’, matTag, height, width, fut, tf, Ife, Ifi, ts, np, ds, Vs, sc, nc, type,
openingArea, openingLength)
This command is used to construct a uniaxialMaterial model that simulates the hysteresis response (Shear
strength-Lateral displacement) of a wood-sheathed cold-formed steel shear wall panel (CFS-SWP). The hystere-
sis model has smooth curves and takes into account the strength and stiffness degradation, as well as pinching
effect.
This uniaxialMaterial gives results in Newton and Meter units, for strength and displacement, respectively.

matTag (int) integer tag identifying material

height (float) SWP’s height (mm)
width (float) SWP’s width (mm)
fut (float) Tensile strength of framing members (MPa)
tf (float) Framing thickness (mm)
Ife (float) Moment of inertia of the double end-stud (mm4)
Ifi (float) Moment of inertia of the intermediate stud (mm4)
ts (float) Sheathing thickness (mm)
np (float) Sheathing number (one or two sides sheathed)
ds (float) Screws diameter (mm)
Vs (float) Screws shear strength (N)
sc (float) Screw spacing on the SWP perimeter (mm)
nc (float) Total number of screws located on the SWP perimeter
type (int) Integer identifier used to define wood sheathing type (DFP=1, OSB=2, CSP=3)
openingArea (float) Total area of openings (mm2)
openingLength (float) Cumulative length of openings (mm)

See also:
Notes

CFSSSWP Steel-Sheathed Cold-formed Steel Shear Wall Panel

uniaxialMaterial(’CFSSSWP’, matTag, height, width, fuf, fyf, tf, Af, fus, fys, ts, np, ds, Vs, sc, dt,
openingArea, openingLength)
This command is used to construct a uniaxialMaterial model that simulates the hysteresis response (Shear
strength-lateral Displacement) of a Steel-Sheathed Cold-Formed Steel Shear Wall Panel (CFS-SWP). The hys-
teresis model has smooth curves and takes into account the strength and stiffness degradation, as well as pinching
effect.
This uniaxialMaterial gives results in Newton and Meter units, for strength and displacement, respectively.

176 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

height (float) SWP’s height (mm)
width (float) SWP’s width (mm)
fuf (float) Tensile strength of framing members (MPa)
fyf (float) Yield strength of framing members (MPa)
tf (float) Framing thickness (mm)
Af (float) Framing cross section area (mm2)
fus (float) Tensile strength of steel sheet sheathing (MPa)
fys (float) Yield strength of steel sheet sheathing (MPa)
ts (float) Sheathing thickness (mm)
np (float) Sheathing number (one or two sides sheathed)
ds (float) Screws diameter (mm)
Vs (float) Screws shear strength (N)
sc (float) Screw spacing on the SWP perimeter (mm)
dt (float) Anchor bolt’s diameter (mm)
openingArea (float) Total area of openings (mm2)
openingLength (float) Cumulative length of openings (mm)

See also:
Notes

Backbone Material

uniaxialMaterial(’Backbone’, matTag, backboneTag)

This command uses a Hysteretic Backbone object to represent a path-independent uniaxial material. Since it is
path-independent, no state information is stored by BackboneMaterial.

matTag (int) integer tag identifying material

backboneTag (int) integer tag identifying a predefined backbone function.

Hysteretic Backbone

HystereticBackbone represents a backbone curve for hysteretic models.

1. ArctangentBackbone
2. BilinearBackbone
3. ManderBackbone
4. MultilinearBackbone
5. TrilinearBackbone
6. MaterialBackbone
7. ReeseStiffClayBelowWS
8. ReeseStiffClayAboveWS
9. VuggyLimestone
10. CementedSoil
11. WeakRock

1.4. Model Commands 177

OpenSeesPy Documentation, Release 3.5.1.3

12. LiquefiedSand

ArctangentBackbone

hystereticBackbone(’Arctangent’, backboneTag, K1, gamma, alpha)

The backbone function 𝐹 (𝑥) is developed by Ranzo and Petrangeli (1998) and defined as

𝐹 (𝑥) = 𝐾1 𝑎𝑡𝑎𝑛(𝐾2 𝑥)
𝐾2 = 𝑡𝑎𝑛(𝛼)/𝛾

backboneTag (int) integer tag identifying the backbone function.

K1 (float) parameter 𝐾1 .
gamma (float) parameter 𝛾.
alpha (float) parameter 𝛼.

BilinearBackbone

hystereticBackbone(’Bilinear’, backboneTag, e1, s1, e2, s2)

The backbone function is bilinear defined by two controlling points (𝑒1 , 𝑠1 ) and (𝑒2 , 𝑠2 ).

backboneTag (int) integer tag identifying the backbone function.

e1 (float) parameter 𝑒1 .
s1 (float) parameter 𝑠1 .
e2 (float) parameter 𝑒2 .
s2 (float) parameter 𝑠2 .

ManderBackbone

hystereticBackbone(’Mander’, backboneTag, fc, epsc, E)

The backbone function 𝐹 (𝑥) is developed by Mander, Priestly, and Park (1988) and defined as
𝑠𝑟
𝐹 (𝑥) = −𝑓𝑐
𝑟 − 1 + 𝑠𝑟
𝑠 = 𝑥/𝜖𝑐
𝐸
𝑟=
𝐸 − 𝑓𝑐 /𝜖𝑐

backboneTag (int) integer tag identifying the backbone function.

fc (float) parameter 𝑓𝑐 .
epsc (float) parameter 𝜖𝑐 .
E (float) parameter 𝐸.

MultilinearBackbone

hystereticBackbone(’Multilinear’, backboneTag, e1, s1, e2, s2, ...)

The backbone function is trilinear defined by any number of controlling points (𝑒1 , 𝑠1 ), (𝑒2 , 𝑠2 ), . . .

178 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

backboneTag (int) integer tag identifying the backbone function.

e1 (float) parameter 𝑒1 .
s1 (float) parameter 𝑠1 .
e2 (float) parameter 𝑒2 .
s2 (float) parameter 𝑠2 .

TrilinearBackbone

hystereticBackbone(’Trilinear’, backboneTag, e1, s1, e2, s2, e3, s3)

The backbone function is trilinear defined by three controlling points (𝑒1 , 𝑠1 ), (𝑒2 , 𝑠2 ), and (𝑒3 , 𝑠3 )

backboneTag (int) integer tag identifying the backbone function.

e1 (float) parameter 𝑒1 .
s1 (float) parameter 𝑠1 .
e2 (float) parameter 𝑒2 .
s2 (float) parameter 𝑠2 .
e3 (float) parameter 𝑒3 .
s3 (float) parameter 𝑠3 .

MaterialBackbone

hystereticBackbone(’Material’, backboneTag, matTag, ’-compression’)

The backbone function treats a uniaxial material as a hysteretic backbone by removing path dependency, i.e.
commitState is never called on the uniaxial material.

backboneTag (int) integer tag identifying the backbone function.

matTag (int) tag for a predefined uniaxial material

ReeseStiffClayBelowWS

hystereticBackbone(’ReeseStiffClayBelowWS’, backboneTag, Esi, y50, As, Pc)

The backbone function is defined in this manual in page 328

backboneTag (int) integer tag identifying the backbone function.

Esi (float) see below
y50 (float) see below
As (float) see below
Pc (float) see below

1.4. Model Commands 179

OpenSeesPy Documentation, Release 3.5.1.3

Esi

𝐸𝑠𝑖 = 𝑘𝑠 𝑥
where ks is the k value for static load selected from the following table
The average shear strength should be computed from the shear strength of the soil to a depth of 5 pile diameters. It
should be defined as half the total maximum principal stress difference in an unconsolidated undrained triaxial test.

Average Undrained Shear Strength (ton/ft^2) 0.5-1 1-2 2-4

𝑘𝑠 (Static) 𝑙𝑏/𝑖𝑛3 500 1000 2000

y50

𝑦50 = 𝜀50 𝑏
Use an appropriate value of 𝜀50 from results of laboratory tests or, in the absence of laboratory tests, from the following
table.

Average Undrained Shear Strength (ton/ft^2) 0.5-1 1-2 2-4

𝜀50 𝑖𝑛/𝑖𝑛 0.007 0.005 0.004

180 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

As

Choose the appropriate value of As from the following figure for the particular nondimensional depth.

1.4. Model Commands 181

OpenSeesPy Documentation, Release 3.5.1.3

182 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Pc

Compute the ultimate soil resistance per unit length of pile, using the smaller of the values given by the equations
below
𝑝𝑐𝑡 = 2𝑐𝑎 𝑏 + 𝛾 ′ 𝑏𝑥 + 2.83𝑐𝑎 𝑥
𝑝𝑐𝑑 = 11𝑐𝑏

where
1. Obtain values for undrained soil shear strength 𝑐, soil submerged unit weight 𝛾 ′ , and pile diameter 𝑏 2. Compute
the average undrained soil shear strength 𝑐𝑎 over the depth 𝑥.

ReeseStiffClayAboveWS

hystereticBackbone(’ReeseStiffClayAboveWS’, backboneTag, pu, y50)

The backbone function is defined in this manual in page 336

backboneTag (int) integer tag identifying the backbone function.

pu (float) see below
y50 (float) see below

1.4. Model Commands 183

OpenSeesPy Documentation, Release 3.5.1.3

pu

Obtain values for undrained shear strength 𝑐, soil unit weight 𝛾, and pile diameter 𝑏. Also obtain the values of 𝜀50
from stress-strain curves. If no stress-strain curves are available, use a value from 𝜀50 of 0.010 or 0.005 as given in the
following table, the larger value being more conservative.

Consistency of Clay 𝜀50

Soft 0.020
Medium 0.010
Stiff 0.005

Compute the ultimate soil resistance per unit length of shaft, 𝑝𝑢 , using the smaller of the values given by the following
equations

𝛾′ 𝐽
𝑝𝑢 = (3 + 𝑥 + 𝑥)𝑐𝑏𝑝𝑢 = 9𝑐𝑏
𝑐 𝑏

y50

Compute the deflection, 𝑦50 at one-half the ultimate soil resistance from the following equation

𝑦50 = 2.5𝜀50 𝑏

VuggyLimestone

hystereticBackbone(’VuggyLimestone’, backboneTag, b, su)

The backbone function is defined in this manual in page 348

backboneTag (int) integer tag identifying the backbone function.

b (float) pile diameter
su (float) shear strength of the rock

184 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

CementedSoil

hystereticBackbone(’CementedSoil’, backboneTag, pM, pU, Kpy, z, b)

The Evans and Duncan (1982) SILT model can be found here

backboneTaginteger tag identifying the backbone function.

(int)
pM (float) soil resistance at point M in figure below
pU (float) Evans and Duncan (1982) proposed a procedure for developing p-y curves and deciding pU,
the ultimate soil resistance, in silty soils, which also defines the point U
Kpy (float) The initial modulus of subgrade reaction defines the initial straight line between O and K in
figure below
z (float) the depth below ground surface to p-y curve.
b (float) diameter (width) of the pile.

1.4. Model Commands 185

OpenSeesPy Documentation, Release 3.5.1.3

There are no available recommendations on developing p-y curves for soil with cohesion and friction angle that are
generally accepted. Generally, in design, the soil is classified in only two different types: cohesive or cohesionless. In
the practice, this simplification leads to a significantly conservative design in the case of cemented soil or silt, which
always neglects the soil resistance from the cohesion component (Juirnarongrit et al. 2005).

The initial straight line (EQ. 21) is defined as

𝑝 = 𝐾𝑝𝑦 𝑧𝑦

the parabolic portion (EQ. 14) is defined as

𝑝 = 𝐶𝑦 1/𝑛

where
𝑝𝑀
𝐶= 1/𝑛
𝑦𝑀
𝑝𝑀
𝑛=
𝑚𝑦𝑀
𝑝𝑈 − 𝑝𝑀
𝑚=
𝑦𝑈 − 𝑦𝑀

The point 𝐾 is defined at 𝑦𝐾 = (𝐶/(𝐾𝑝𝑦 𝑧))𝑛/(𝑛−1) and 𝑝𝐾 = 𝑦𝐾 𝐾𝑝𝑦 .

The point 𝑀 is defined at 𝑦𝑀 = 𝑏/60 and 𝑝𝑀 .
The point 𝑈 is defined at 𝑦𝑈 = 3𝑏/80 and 𝑝𝑈 .

186 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

WeakRock

hystereticBackbone(’WeakRock’, backboneTag, Kir, pur, yrm)

The backbone function is defined on page 56 of this manual

backboneTag (int) integer tag identifying the backbone function.

Kir (float) see below
pur (float) see below
yrm (float) see below

For “weak rock” defined as rock with unconfined compressive strength between 0.5 MPa and 5 MPa, the shape of the
p-y curve, as shown in above Figure, can be described by the following equations. For the initial linear portion of the
curve

𝑝 = 𝐾𝑖𝑟 𝑦, for 𝑦 ≤ 𝑦𝐴

For the transitional and nonlinear portion

(︂ )︂0.25
𝑝𝑢𝑟 𝑦
𝑝= , for 𝑦 ≥ 𝑦𝐴 , 𝑝 ≤ 𝑝𝑢𝑟
2 𝑦𝑟𝑚

and when the ultimate resistance is reached

1.4. Model Commands 187

OpenSeesPy Documentation, Release 3.5.1.3

Kir

𝐾𝑖𝑟 is the initial slope of the curve, which is related to the initial elastic modulus of the rock mass as follows

𝐾𝑖𝑟 u 𝑘𝑖𝑟 𝐸𝑖𝑟

where 𝐸𝑖𝑟 = rock mass initial elastic modulus and 𝑘𝑖𝑟 = dimensionless constant given by
(︂ )︂
400𝑥𝑟
𝑘𝑖𝑟 = 100 + , for 0 ≤ 𝑥𝑟 ≤ 3𝐵
3𝐵
𝑘𝑖𝑟 = 500, for 𝑥𝑟 ≥ 3𝐵

where 𝐵 is shaft diameter, 𝑥𝑟 is the depth below rock surface.

pur

𝑝𝑢𝑟 is the rock mass ultimate resistance defined as

(︁ 𝑥𝑟 )︁
𝑝𝑢𝑟 = 𝛼𝑟 𝑞𝑢 𝐵 1 + 1.4 , for 0 ≤ 𝑥𝑟 ≤ 3𝐵
𝐵
𝑝𝑢𝑟 = 5.2𝛼𝑟 𝑞𝑢 𝐵, for 𝑥𝑟 ≥ 3𝐵

in which 𝑞𝑢 = uniaxial compressive strength of intact rock, 𝛼𝑟 = strength reduction factor, and 𝑥𝑟 = depth below rock
surface. Selection of 𝛼𝑟 is based on the assumption that fracturing will occur at the surface of the rock under small
deflections, thus reducing the rock mass compressive strength. The value of 𝛼𝑟 is assumed to be one-third for RQD
of 100 and to increase linearly to unity at RQD of zero. The underlying assumption is that, if the rock mass is already
highly fractured, then no additional fracturing with accompanying strength loss will occur. However, this approach
appears to have a fundamental shortcoming in that it relies on the compressive strength of the intact rock and not the
strength of the rock mass. For a highly fractured rock mass (low RQD) with a high-intact rock strength, it seems that
the rock mass strength could be overestimated.

yrm

𝑦𝑟𝑚 = 𝑘𝑟𝑚 𝐵

where
𝑘𝑟𝑚 is a constant ranging from 0.0005 to 0.00005 that serves to establish the overall stiffness of the curve.
𝐵 is the shaft diameter.

LiquefiedSand

hystereticBackbone(’LiquefiedSand’, backboneTag, X, D, kN, m)

The p-y curve model of liquefied sand is recommended in Rollins (2005) based on the full-scale lateral load
tests to analyze the laterally loaded single piles when the liquefied sand layer is present.
The backbone function is defined in manual

188 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

backboneTag integer tag identifying the backbone function.

(int)
X (float) the depth.
D (float) is the pile diameter
kN (float) Since the model assumes the metric units, if your units are different to, define kN to
match your unit system.
m (float) Define the unit meter for the same reason of kN

The typical p-y curve for liquefied sand is shown in the figure above. The pile lateral soil resistance, P, increases with
the pile deflection along a concave upward curve until the deflection reaches 150 mm. Note that 𝑃𝑢 is the ultimate
lateral resistance when the pile deflection (y value) is equal to or greater than 150 mm.
The following equations are proposed by Rollins et al. (2005) to estimate the mobilization of pile lateral soil resistance
(P) with deflection (y):

𝑃0.3𝑚 = 𝐴(𝐵𝑦)𝐶 ≤ 15𝑘𝑁/𝑚

𝑃𝑑 = 3.81 ln |𝐷| + 5.6, for 0.3𝑚𝐷2.6𝑚
𝑃 = 𝑃0.3𝑚 𝑃𝑑
−7
𝐴 = 3 × 10 (𝑋 + 1)6.05
𝐵 = 2.8(𝑋 + 1)0.11
𝐶 = 2.85(𝑋 + 1)−0.41

Note that A, B, and C in the equations above are the curve fitting parameters that depend on the depth (X). D is the
pile diameter. 𝑃𝑑 is the pile diameter correction factor and it is limited to the pile diameter between 0.3 m and 2.6 m.

1.4. Model Commands 189

OpenSeesPy Documentation, Release 3.5.1.3

1.4.15 nDMaterial commands

nDMaterial(matType, matTag, *matArgs)

This command is used to construct an NDMaterial object which represents the stress-strain relationship at the
gauss-point of a continuum element.

matType (str) material type

matTag (int) material tag.
matArgs (list) a list of material arguments, must be preceded with *.

For example,

matType = 'ElasticIsotropic'
matTag = 1
matArgs = [E, v]
nDMaterial(matType, matTag, *matArgs)

Standard Models

The following contain information about available matType:

1. ElasticIsotropic
2. ElasticOrthotropic
3. J2Plasticity
4. DruckerPrager
5. PlaneStress
6. PlaneStrain
7. MultiaxialCyclicPlasticity
8. BoundingCamClay
9. PlateFiber
10. FSAM
11. ManzariDafalias
12. PM4Sand
13. PM4Silt
14. StressDensityModel
15. AcousticMedium

ElasticIsotropic

nDMaterial(’ElasticIsotropic’, matTag, E, nu, rho=0.0)

This command is used to construct an ElasticIsotropic material object.

190 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

E (float) elastic modulus
nu (float) Poisson’s ratio
rho (float) mass density (optional)

The material formulations for the ElasticIsotropic object are:

• 'ThreeDimensional'
• 'PlaneStrain'
• 'Plane Stress'
• 'AxiSymmetric'
• 'PlateFiber'

ElasticOrthotropic

nDMaterial(’ElasticOrthotropic’, matTag, Ex, Ey, Ez, nu_xy, nu_yz, nu_zx, Gxy, Gyz, Gzx, rho=0.0)
This command is used to construct an ElasticOrthotropic material object.

matTag (int) integer tag identifying material

Ex (float) elastic modulus in x direction
Ey (float) elastic modulus in y direction
Ez (float) elastic modulus in z direction
nu_xy (float) Poisson’s ratios in x and y plane
nu_yz (float) Poisson’s ratios in y and z plane
nu_zx (float) Poisson’s ratios in z and x plane
Gxy (float) shear modulii in x and y plane
Gyz (float) shear modulii in y and z plane
Gzx (float) shear modulii in z and x plane
rho (float) mass density (optional)

The material formulations for the ElasticOrthotropic object are:

• 'ThreeDimensional'
• 'PlaneStrain'
• 'Plane Stress'
• 'AxiSymmetric'
• 'BeamFiber'
• 'PlateFiber'

J2Plasticity

nDMaterial(’J2Plasticity’, matTag, K, G, sig0, sigInf, delta, H)

This command is used to construct an multi dimensional material object that has a von Mises (J2) yield criterium
and isotropic hardening.

1.4. Model Commands 191

OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

K (float) bulk modulus
G (float) shear modulus
sig0 (float) initial yield stress
sigInf (float) final saturation yield stress
delta (float) exponential hardening parameter
H (float) linear hardening parameter

The material formulations for the J2Plasticity object are:

• 'ThreeDimensional'
• 'PlaneStrain'
• 'Plane Stress'
• 'AxiSymmetric'
• 'PlateFiber'
J2 isotropic hardening material class
Elastic Model

𝜎 = 𝐾 * 𝑡𝑟𝑎𝑐𝑒(𝜖𝑒 ) + (2 * 𝐺) * 𝑑𝑒𝑣(𝜖𝑒 )

Yield Function
√︀
𝜑(𝜎, 𝑞) = ||𝑑𝑒𝑣(𝜎)|| − ( 23 * 𝑞(𝑥𝑖 ))

Saturation Isotropic Hardening with linear term

𝑞(𝑥𝑖 ) = 𝜎0 + (𝜎∞ − 𝜎0 ) * 𝑒𝑥𝑝(−𝑑𝑒𝑙𝑡𝑎 * 𝜉) + 𝐻 * 𝜉

Flow Rules
𝜕𝜑
𝜖˙𝑝 = 𝛾 *
𝜕𝜎
𝜕𝜑
𝜉˙ = −𝛾 *
𝜕𝑞
Linear Viscosity

𝜑
𝛾= (𝑖𝑓 𝜑 > 0)
𝜂
Backward Euler Integration Routine Yield condition enforced at time n+1
set 𝜂 = 0 for rate independent case

DruckerPrager

nDMaterial(’DruckerPrager’, matTag, K, G, sigmaY, rho, rhoBar, Kinf, Ko, delta1, delta2, H, theta, density,
atmPressure=101e3)
This command is used to construct an multi dimensional material object that has a Drucker-Prager yield cri-
terium.

192 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

K (float) bulk modulus
G (float) shear modulus
sigmaY (float) yield stress
rho (float) frictional strength parameter
rhoBar (float) controls evolution of plastic volume change, 0 ≤ 𝑟ℎ𝑜𝐵𝑎𝑟 ≤ 𝑟ℎ𝑜.
Kinf (float) nonlinear isotropic strain hardening parameter, 𝐾𝑖𝑛𝑓 ≥ 0.
Ko (float) nonlinear isotropic strain hardening parameter, 𝐾𝑜 ≥ 0.
delta1 (float) nonlinear isotropic strain hardening parameter, 𝑑𝑒𝑙𝑡𝑎1 ≥ 0.
delta2 (float) tension softening parameter, 𝑑𝑒𝑙𝑡𝑎2 ≥ 0.
H (float) linear hardening parameter, 𝐻 ≥ 0.
theta (float) controls relative proportions of isotropic and kinematic hardening, 0 ≤ 𝑡ℎ𝑒𝑡𝑎 ≤ 1.
density (float) mass density of the material
atmPressure (float) optional atmospheric pressure for update of elastic bulk and shear moduli

The material formulations for the DrukerPrager object are:

• 'ThreeDimensional'
• 'PlaneStrain'
See theory.

PlaneStress

nDMaterial(’PlaneStress’, matTag, mat3DTag)

This command is used to construct a plane-stress material wrapper which converts any three-dimensional mate-
rial into a plane stress material via static condensation.

matTag (int) integer tag identifying material

mat3DTag (int) tag of perviously defined 3d ndMaterial material

The material formulations for the PlaneStress object are:

• 'Plane Stress'

PlaneStrain

nDMaterial(’PlaneStrain’, matTag, mat3DTag)

This command is used to construct a plane-stress material wrapper which converts any three-dimensional mate-
rial into a plane strain material by imposing plain strain conditions on the three-dimensional material.

matTag (int) integer tag identifying material

mat3DTag (int) integer tag of previously defined 3d ndMaterial material

The material formulations for the PlaneStrain object are:

• 'PlaneStrain'

1.4. Model Commands 193

OpenSeesPy Documentation, Release 3.5.1.3

MultiaxialCyclicPlasticity

nDMaterial(’MultiaxialCyclicPlasticity’, matTag, rho, K, G, Su, Ho, h, m, beta, KCoeff )

This command is used to construct an multiaxial Cyclic Plasticity model for clays

matTag (int) integer tag identifying material

rho (float) density
K (float) buck modulus
G (float) maximum (small strain) shear modulus √︀
Su (float) undrained shear strength, size of bounding surface 𝑅 = 8/3 * 𝑆𝑢
Ho (float) linear kinematic hardening modulus of bounding surface
h (float) hardening parameter
m (float) hardening parameter
beta (float) integration parameter, usually beta=0.5
KCoeff (float) coefficient of earth pressure, K0

BoundingCamClay

nDMaterial(’BoundingCamClay’, matTag, massDensity, C, bulkMod, OCR, mu_o, alpha, lambda, h, m)

This command is used to construct a multi-dimensional bounding surface Cam Clay material object after Borja
et al. (2001).

matTag (int) integer tag identifying material

massDensity mass density
(float)
C (float) ellipsoidal axis ratio (defines shape of ellipsoidal loading/bounding surfaces)
bulkMod (float) initial bulk modulus
OCR (float) overconsolidation ratio
mu_o (float) initial shear modulus
alpha (float) pressure-dependency parameter for modulii (greater than or equal to zero)
lambda (float) soil compressibility index for virgin loading
h (float) hardening parameter for plastic response inside of bounding surface (if h = 0, no hard-
ening)
m (float) hardening parameter (exponent) for plastic response inside of bounding surface (if m =
0, only linear hardening)

The material formulations for the BoundingCamClay object are:

• 'ThreeDimensional'
• 'PlaneStrain'
See also for information

PlateFiber

nDMaterial(’PlateFiber’, matTag, threeDTag)

This command is used to construct a plate-fiber material wrapper which converts any three-dimensional material
into a plate fiber material (by static condensation) appropriate for shell analysis.

194 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

threeDTag (float) material tag for a previously-defined three-dimensional material

FSAM

nDMaterial(’FSAM’, matTag, rho, sXTag, sYTag, concTag, rouX, rouY, nu, alfadow)
This command is used to construct a nDMaterial FSAM (Fixed-Strut-Angle-Model, Figure 1, Kolozvari et
al., 2015), which is a plane-stress constitutive model for simulating the behavior of RC panel elements under
generalized, in-plane, reversed-cyclic loading conditions (Ulugtekin, 2010; Orakcal et al., 2012). In the FSAM
constitutive model, the strain fields acting on concrete and reinforcing steel components of a RC panel are
assumed to be equal to each other, implying perfect bond assumption between concrete and reinforcing steel
bars. While the reinforcing steel bars develop uniaxial stresses under strains in their longitudinal direction, the
behavior of concrete is defined using stress-strain relationships in biaxial directions, the orientation of which
is governed by the state of cracking in concrete. Although the concrete stress-strain relationship used in the
FSAM is fundamentally uniaxial in nature, it also incorporates biaxial softening effects including compression
softening and biaxial damage. For transfer of shear stresses across the cracks, a friction-based elasto-plastic
shear aggregate interlock model is adopted, together with a linear elastic model for representing dowel action
on the reinforcing steel bars (Kolozvari, 2013). Note that FSAM constitutive model is implemented to be used
with Shear-Flexure Interaction model for RC walls (SFI_MVLEM), but it could be also used elsewhere.

matTag (int) integer tag identifying material

rho (float) Material density
sXTag (int) Tag of uniaxialMaterial simulating horizontal (x) reinforcement
sYTag (int) Tag of uniaxialMaterial simulating vertical (y) reinforcement
concTag (int) Tag of uniaxialMaterial simulating concrete, shall be used with uniaxialMaterial Con-
creteCM
rouX (float) Reinforcing ratio in horizontal (x) direction (𝑟𝑜𝑢𝑋 =𝑠,𝑥 /𝐴𝑔𝑟𝑜𝑠𝑠,𝑥 )
rouY (float) Reinforcing ratio in vertical (x) direction (𝑟𝑜𝑢𝑌 =𝑠,𝑦 /𝐴𝑔𝑟𝑜𝑠𝑠,𝑦 )
nu (float) Concrete friction coefficient (0.0 < 𝜈 < 1.5)
alfadow Stiffness coefficient of reinforcement dowel action (0.0 < 𝑎𝑙𝑓 𝑎𝑑𝑜𝑤 < 0.05)
(float)

See also here

References:
1) Kolozvari K., Orakcal K., and Wallace J. W. (2015). “Shear-Flexure Interaction Modeling of reinforced Con-
crete Structural Walls and Columns under Reversed Cyclic Loading”, Pacific Earthquake Engineering Research
Center, University of California, Berkeley, PEER Report No. 2015/12
2) Kolozvari K. (2013). “Analytical Modeling of Cyclic Shear-Flexure Interaction in Reinforced Concrete Struc-
tural Walls”, PhD Dissertation, University of California, Los Angeles.
3) Orakcal K., Massone L.M., and Ulugtekin D. (2012). “Constitutive Modeling of Reinforced Concrete Panel
Behavior under Cyclic Loading”, Proceedings, 15th World Conference on Earthquake Engineering, Lisbon,
Portugal.
4) Ulugtekin D. (2010). “Analytical Modeling of Reinforced Concrete Panel Elements under Reversed Cyclic
Loadings”, M.S. Thesis, Bogazici University, Istanbul, Turkey.

1.4. Model Commands 195

OpenSeesPy Documentation, Release 3.5.1.3

ManzariDafalias

nDMaterial(’ManzariDafalias’, matTag, G0, nu, e_init, Mc, c, lambda_c, e0, ksi, P_atm, m, h0, ch, nb, A0,
nd, z_max, cz, Den)
This command is used to construct a multi-dimensional Manzari-Dafalias(2004) material.

matTag (int) integer tag identifying material

G0 (float) shear modulus constant
nu (float) poisson ratio
e_init (float) initial void ratio
Mc (float) critical state stress ratio
c (float) ratio of critical state stress ratio in extension and compression
lambda_c (float) critical state line constant
e0 (float) critical void ratio at p = 0
ksi (float) critical state line constant
P_atm (float) atmospheric pressure
m (float) yield surface constant (radius of yield surface in stress ratio space)
h0 (float) constant parameter
ch (float) constant parameter
nb (float) bounding surface parameter, 𝑛𝑏 ≥ 0
A0 (float) dilatancy parameter
nd (float) dilatancy surface parameter 𝑛𝑑 ≥ 0
z_max (float) fabric-dilatancy tensor parameter
cz (float) fabric-dilatancy tensor parameter
Den (float) mass density of the material

The material formulations for the ManzariDafalias object are:

• 'ThreeDimensional'
• 'PlaneStrain'
See also here
References
Dafalias YF, Manzari MT. “Simple plasticity sand model accounting for fabric change effects”. Journal of Engineering
Mechanics 2004

PM4Sand

nDMaterial(’PM4Sand’, matTag, D_r, G_o, h_po, Den, P_atm, h_o, e_max, e_min, n_b, n_d, A_do, z_max,
c_z, c_e, phi_cv, nu, g_degr, c_dr, c_kaf, Q_bolt, R_bolt, m_par, F_sed, p_sed)
This command is used to construct a 2-dimensional PM4Sand material.

196 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
D_r (float) Relative density, in fraction
G_o (float) Shear modulus constant
h_po Contraction rate parameter
(float)
Den (float) Mass density of the material
P_atm Optional, Atmospheric pressure
(float)
h_o (float) Optional, Variable that adjusts the ratio of plastic modulus to elastic modulus
e_max Optional, Maximum and minimum void ratios
(float)
e_min Optional, Maximum and minimum void ratios
(float)
n_b (float) Optional, Bounding surface parameter, 𝑛𝑏 ≥ 0
n_d (float) Optional, Dilatancy surface parameter 𝑛𝑑 ≥ 0
A_do Optional, Dilatancy parameter, will be computed at the time of initialization if input value is
(float) negative
z_max Optional, Fabric-dilatancy tensor parameter
(float)
c_z (float) Optional, Fabric-dilatancy tensor parameter
c_e (float) Optional, Variable that adjusts the rate of strain accumulation in cyclic loading
phi_cv Optional, Critical state effective friction angle
(float)
nu (float) Optional, Poisson’s ratio
g_degr Optional, Variable that adjusts degradation of elastic modulus with accumulation of fabric
(float)
c_dr Optional, Variable that controls the rotated dilatancy surface
(float)
c_kaf Optional, Variable that controls the effect that sustained static shear stresses have on plastic
(float) modulus
Q_bolt Optional, Critical state line parameter
(float)
R_bolt Optional, Critical state line parameter
(float)
m_par Optional, Yield surface constant (radius of yield surface in stress ratio space)
(float)
F_sed Optional, Variable that controls the minimum value the reduction factor of the elastic moduli
(float) can get during reconsolidation
p_sed Optional, Mean effective stress up to which reconsolidation strains are enhanced
(float)

The material formulations for the PM4Sand object are:

• 'PlaneStrain'
See als here
References
R.W.Boulanger, K.Ziotopoulou. “PM4Sand(Version 3.1): A Sand Plasticity Model for Earthquake Engineering Ap-
plications”. Report No. UCD/CGM-17/01 2017

1.4. Model Commands 197

OpenSeesPy Documentation, Release 3.5.1.3

PM4Silt

nDMaterial(’PM4Silt’, matTag, S_u, Su_Rat, G_o, h_po, Den <Su_factor, Patm, nu, nG, h0, eInit, lambda,
phicv, nb_wet, nb_dry, nd, Ado, ru_max, zmax, cz, ce, Cgd, ckaf, m_m, CG_consol>)
Code Developed by: Long Chen and pedro at U.Washington.
[Boulanger-Ziotopoulou2018]
This command is used to construct a 2-dimensional PM4Silt material.

matTag integer tag identifying material

(int)
S_u (float) Undrained shear strength
Su_Rat Undrained shear strength ratio.
(float)
G_o (float) Shear modulus constant
h_po (float) Contraction rate parameter
Den (float) Mass density of the material
Su_factor Optional: Undrained shear strength reduction factor
(float)
P_atm Optional: Atmospheric pressure
(float)
nu (float) Optional: Poisson’s ratio. Default value is 0.3.
nG (float) Optional: Shear modulus exponent. Default value is 0.75.
h0 (float) Optional: Variable that adjusts the ratio of plastic modulus to elastic modulus. Default value
is 0.5.
eInit Optional: Initial void ratios. Default value is 0.90.
(float)
lambda Optional: The slope of critical state line in e-ln(p) space. Default value is 0.060.
(float)
phicv Optional: Critical state effective friction angle. Default value is 32 degrees.
(float)
nb_wet Optional: Bounding surface parameter for loose of critical state conditions 1.0 ≥ 𝑛𝑏𝑤 𝑒𝑡 ≥
(float) 0.01. Default value is 0.8. in cyclic loading
nb_dry Optional: Bounding surface parameter for dense of critical state conditions 𝑛𝑏𝑑 𝑟𝑦 ≥ 0.
(float) Default value is 0.5.
nd (float) Optional: Dilatancy surface parameter 𝑛𝑑 ≥ 0. Default value is 0.3.
Ado (float) Optional: Dilatancy parameter. Default value is 0.8. with accumulation of fabric
ru_max Optional: Maximum pore pressure ratio based on p’.
(float)
z_max Optional: Fabric-dilatancy tensor parameter
(float)
cz (float) Optional: Fabric-dilatancy tensor parameter. Default value is 100.0.
ce (float) Optional: Variable that adjusts the rate of strain accumulation in cyclic loading
cgd (float) Optional: Variable that adjusts degradation of elastic modulus with accumulation of fabric.
Default value is 3.0. in stress ratio space)
ckaf (float) Optional: Variable that controls the effect that sustained static shear stresses have on plastic
modulus. Default value is 4.0.
m_m (float) Optional: Yield surface constant (radius of yield surface in stress ratio space). Default value
is 0.01.
CG_consol Optional: Reduction factor of elastic modulus for reconsolidation. 𝐶𝐺𝑐 𝑜𝑛𝑠𝑜𝑙 ≥ 1. Default
(float) value is 2.0.

198 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Note: The material formulation for the PM4Silt object is PlaneStrain

If both S_u and Su_Rat values are specified the value of S_u is used.
Valid Element Recorder queries are stress, strain, alpha (or backstressratio) for 𝛼, fabric for z, and alpha_in (or
alphain) for 𝛼in
Elastic or response could be enforced by

updateMaterialStage('-material', matTag, '-stage', 0)

Elastoplastic by

updateMaterialStage('-material', matTag, '-stage', 1)

The program will use the default value of a secondary parameter if a negative input is assigned to that parameter, e.g.
Ado = -1. However, FirstCall is mandatory when switching from elastic to elastoplastic if negative inputs are assigned
to stress-dependent secondary parameters, e.g. Ado and zmax. FirstCall can be set as,

setParameter('-value', 0, '-ele', elementTag, 'FirstCall', matTag)

Post-shake reconsolidation can be activated by

setParameter('-value', 1, '-ele', elementTag, 'Postshake', matTag)

The user should check that the results are not sensitive to time step size.

Refer to PM4Sand for examples.

StressDensityModel

nDMaterial(’stressDensity’, matTag, mDen, eNot, A, n, nu, a1, b1, a2, b2, a3, b3, fd, muNot, muCyc, sc, M,
patm, *ssls, hsl, p1)
This command is used to construct a multi-dimensional stress density material object for modeling sand be-
haviour following the work of Cubrinovski and Ishihara (1998a,b).

1.4. Model Commands 199

OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
mDen mass density
(float)
eNot initial void ratio
(float)
A (float) constant for elastic shear modulus
n (float) pressure dependency exponent for elastic shear modulus
nu (float) Poisson’s ratio
a1 (float) peak stress ratio coefficient (𝑒𝑡𝑎𝑀 𝑎𝑥 = 𝑎1 + 𝑏1 * 𝐼𝑠)
b1 (float) peak stress ratio coefficient (𝑒𝑡𝑎𝑀 𝑎𝑥 = 𝑎1 + 𝑏1 * 𝐼𝑠)
a2 (float) max shear modulus coefficient (𝐺𝑛𝑚 𝑎𝑥 = 𝑎2 + 𝑏2 * 𝐼𝑠)
b2 (float) max shear modulus coefficient (𝐺𝑛𝑚 𝑎𝑥 = 𝑎2 + 𝑏2 * 𝐼𝑠)
a3 (float) min shear modulus coefficient (𝐺𝑛𝑚 𝑖𝑛 = 𝑎3 + 𝑏3 * 𝐼𝑠)
b3 (float) min shear modulus coefficient (𝐺𝑛𝑚 𝑖𝑛 = 𝑎3 + 𝑏3 * 𝐼𝑠)
fd (float) degradation constant
muNot dilatancy coefficient (monotonic loading)
(float)
muCyc dilatancy coefficient (cyclic loading)
(float)
sc (float) dilatancy strain
M (float) critical state stress ratio
patm atmospheric pressure (in appropriate units)
(float)
ssls void ratio of quasi steady state (QSS-line) at pressures [pmin, 10kPa, 30kPa, 50kPa, 100kPa,
(list 200kPa, 400kPa] (default = [0.877, 0.877, 0.873, 0.870, 0.860, 0.850, 0.833])
(float))
hsl void ratio of upper reference state (UR-line) for all pressures (default = 0.895)
(float)
p1 (float) pressure corresponding to ssl1 (default = 1.0 kPa)

The material formulations for the StressDensityModel object are:

• 'ThreeDimensional'
• 'PlaneStrain'
References
Cubrinovski, M. and Ishihara K. (1998a) ‘Modelling of sand behaviour based on state concept,’ Soils and Foundations,
38(3), 115-127.
Cubrinovski, M. and Ishihara K. (1998b) ‘State concept and modified elastoplasticity for sand modelling,’ Soils and
Foundations, 38(4), 213-225.
Das, S. (2014) Three Dimensional Formulation for the Stress-Strain-Dilatancy Elasto-Plastic Constitutive Model for
Sand Under Cyclic Behaviour, Master’s Thesis, University of Canterbury.

AcousticMedium

nDMaterial(’AcousticMedium’, matTag, K, rho)

This command is used to construct an acoustic medium NDMaterial object.

200 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag (int) integer tag identifying material

K (float) bulk module of the acoustic medium
rho (float) mass density of the acoustic medium

Tsinghua Sand Models

1. CycLiqCP
2. CycLiqCPSP

CycLiqCP

nDMaterial(’CycLiqCP’, matTag, G0, kappa, h, Mfc, dre1, Mdc, dre2, rdr, alpha, dir, ein, rho)
This command is used to construct a multi-dimensional material object that that follows the constitutive behavior
of a cyclic elastoplasticity model for large post- liquefaction deformation.
CycLiqCP material is a cyclic elastoplasticity model for large post-liquefaction deformation, and is implemented
using a cutting plane algorithm. The model is capable of reproducing small to large deformation in the pre- to
post-liquefaction regime. The elastic moduli of the model are pressure dependent. The plasticity in the model
is developed within the framework of bounding surface plasticity, with special consideration to the formulation
of reversible and irreversible dilatancy.
The model does not take into consideration of the state of sand, and requires different parameters for sand under
different densities and confining pressures. The surfaces (i.e. failure and maximum pre-stress) are considered as
circles in the pi plane.
The model has been validated against VELACS centrifuge model tests and has used on numerous simulations of
liquefaction related problems.
When this material is employed in regular solid elements (e.g., FourNodeQuad, Brick), it simulates drained soil
response. When solid-fluid coupled elements (u-p elements and SSP u-p elements) are used, the model is able to
simulate undrained and partially drained behavior of soil.

matTag (int) integer tag identifying material

G0 (float) A constant related to elastic shear modulus
kappa (float) bulk modulus
h (float) Model parameter for plastic modulus
Mfc (float) Stress ratio at failure in triaxial compression
dre1 (float) Coefficient for reversible dilatancy generation
Mdc (float) Stress ratio at which the reversible dilatancy sign changes
dre2 (float) Coefficient for reversible dilatancy release
rdr (float) Reference shear strain length
alpha (float) Parameter controlling the decrease rate of irreversible dilatancy
dir (float) Coefficient for irreversible dilatancy potential
ein (float) Initial void ratio
rho (float) Saturated mass density

The material formulations for the CycLiqCP object are:

• 'ThreeDimensional'
• 'PlaneStrain'
See also here

1.4. Model Commands 201

OpenSeesPy Documentation, Release 3.5.1.3

CycLiqCPSP

nDMaterial(’CycLiqCPSP’, matTag, G0, kappa, h, M, dre1, dre2, rdr, alpha, dir, lambdac, ksi, e0, np, nd,
ein, rho)
This command is used to construct a multi-dimensional material object that that follows the constitutive behavior
of a cyclic elastoplasticity model for large post- liquefaction deformation.
CycLiqCPSP material is a constitutive model for sand with special considerations for cyclic behaviour and
accumulation of large post-liquefaction shear deformation, and is implemented using a cutting plane algorithm.
The model: (1) achieves the simulation of post-liquefaction shear deformation based on its physics, allowing the
unified description of pre- and post-liquefaction behavior of sand; (2) directly links the cyclic mobility of sand
with reversible and irreversible dilatancy, enabling the unified description of monotonic and cyclic loading; (3)
introduces critical state soil mechanics concepts to achieve unified modelling of sand under different states.
The critical, maximum stress ratio and reversible dilatancy surfaces follow a rounded triangle in the pi plane similar
to the Matsuoka-Nakai criterion.
When this material is employed in regular solid elements (e.g., FourNodeQuad, Brick), it simulates drained soil
response. When solid-fluid coupled elements (u-p elements and SSP u-p elements) are used, the model is able to
simulate undrained and partially drained behavior of soil.

matTag (int) integer tag identifying material

G0 (float) A constant related to elastic shear modulus
kappa (float) bulk modulus
h (float) Model parameter for plastic modulus
M (float) Critical state stress ratio
dre1 (float) Coefficient for reversible dilatancy generation
dre2 (float) Coefficient for reversible dilatancy release
rdr (float) Reference shear strain length
alpha (float) Parameter controlling the decrease rate of irreversible dilatancy
dir (float) Coefficient for irreversible dilatancy potential
lambdac (float) Critical state constant
ksi (float) Critical state constant
e0 (float) Void ratio at pc=0
np (float) Material constant for peak mobilized stress ratio
nd (float) Material constant for reversible dilatancy generation stress ratio
ein (float) Initial void ratio
rho (float) Saturated mass density

The material formulations for the CycLiqCP object are:

• 'ThreeDimensional'
• 'PlaneStrain'
See also here
REFERENCES: Wang R., Zhang J.M., Wang G., 2014. A unified plasticity model for large post-liquefaction shear
deformation of sand. Computers and Geotechnics. 59, 54-66.

Materials for Modeling Concrete Walls

1. PlaneStressUserMaterial
2. PlateFromPlaneStress
3. PlateRebar

202 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

4. PlasticDamageConcretePlaneStress

PlaneStressUserMaterial

nDMaterial(’PlaneStressUserMaterial’, matTag, nstatevs, nprops, fc, ft, fcu, epsc0, epscu, epstu, stc)
This command is used to create the multi-dimensional concrete material model that is based on the damage
mechanism and smeared crack model.

nstatevs (int) number of state/history variables (usually 40)

nprops (int) number of material properties (usually 7)
matTag (int) integer tag identifying material
fc (float) concrete compressive strength at 28 days (positive)
ft (float) concrete tensile strength (positive)
fcu (float) concrete crushing strength (negative)
epsc0 (float) concrete strain at maximum strength (negative)
epscu (float) concrete strain at crushing strength (negative)
epstu (float) ultimate tensile strain (positive)
stc (float) shear retention factor

PlateFromPlaneStress

nDMaterial(’PlateFromPlaneStress’, matTag, pre_def_matTag, OutofPlaneModulus)

This command is used to create the multi-dimensional concrete material model that is based on the damage
mechanism and smeared crack model.

matTag (int) new integer tag identifying material deriving from pre-defined PlaneStress
material
pre_def_matTag (int) integer tag identifying PlaneStress material
OutofPlaneModulus shear modulus for out of plane stresses
(float)

PlateRebar

nDMaterial(’PlateRebar’, matTag, pre_def_matTag, sita)

This command is used to create the multi-dimensional reinforcement material.

matTag (int) new integer tag identifying material deriving from pre-defined uniaxial material
pre_def_matTag (int) integer tag identifying uniaxial material
sita (float) define the angle of reinforcement layer, 90 (longitudinal), 0 (tranverse)

PlasticDamageConcretePlaneStress

nDMaterial(’PlasticDamageConcretePlaneStress’, matTag, E, nu, ft, fc, <beta, Ap, An, Bn>)

No documentation is available yet. If you have the manual, please let me know.

1.4. Model Commands 203

OpenSeesPy Documentation, Release 3.5.1.3

Contact Materials for 2D and 3D

1. ContactMaterial2D
2. ContactMaterial3D

ContactMaterial2D

nDMaterial(’ContactMaterial2D’, matTag, mu, G, c, t)

This command is used to construct a ContactMaterial2D nDMaterial object.

matTag (int) integer tag identifying material

mu (float) interface frictional coefficient
G (float) interface stiffness parameter
c (float) interface cohesive intercept
t (float) interface tensile strength

The ContactMaterial2D nDMaterial defines the constitutive behavior of a frictional interface between two bodies in
contact. The interface defined by this material object allows for sticking, frictional slip, and separation between the
two bodies in a two-dimensional analysis. A regularized Coulomb frictional law is assumed. Information on the theory
behind this material can be found in, e.g. Wriggers (2002).

Note:
1. The ContactMaterial2D nDMaterial has been written to work with the SimpleContact2D and BeamContact2D
element objects.
2. There are no valid recorder queries for this material other than those which are listed with those elements

References:
Wriggers, P. (2002). Computational Contact Mechanics. John Wilely & Sons, Ltd, West Sussex, England.

ContactMaterial3D

nDMaterial(’ContactMaterial3D’, matTag, mu, G, c, t)

This command is used to construct a ContactMaterial3D nDMaterial object.

matTag (int) integer tag identifying material

mu (float) interface frictional coefficient
G (float) interface stiffness parameter
c (float) interface cohesive intercept
t (float) interface tensile strength

The ContactMaterial3D nDMaterial defines the constitutive behavior of a frictional interface between two bodies in
contact. The interface defined by this material object allows for sticking, frictional slip, and separation between the
two bodies in a three-dimensional analysis. A regularized Coulomb frictional law is assumed. Information on the
theory behind this material can be found in, e.g. Wriggers (2002).

Note:

204 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1. The ContactMaterial3D nDMaterial has been written to work with the SimpleContact3D and BeamContact3D
element objects.
2. There are no valid recorder queries for this material other than those which are listed with those elements.

References:
Wriggers, P. (2002). Computational Contact Mechanics. John Wilely & Sons, Ltd, West Sussex, England.

Wrapper material for Initial State Analysis

1. InitialStateAnalysisWrapper
2. Initial Stress Material
3. Initial Strain Material

InitialStateAnalysisWrapper

nDMaterial(’InitialStateAnalysisWrapper’, matTag, nDMatTag, nDim)

The InitialStateAnalysisWrapper nDMaterial allows for the use of the InitialStateAnalysis command for setting
initial conditions. The InitialStateAnalysisWrapper can be used with any nDMaterial. This material wrapper
allows for the development of an initial stress field while maintaining the original geometry of the problem. An
example analysis is provided below to demonstrate the use of this material wrapper object.

matTag (int) integer tag identifying material

nDMatTag (int) the tag of the associated nDMaterial object
nDim (int) number of dimensions (2 for 2D, 3 for 3D)

Note:
1. There are no valid recorder queries for the InitialStateAnalysisWrapper.
2. The InitialStateAnalysis off command removes all previously defined recorders. Two sets of recorders are
needed if the results before and after this command are desired. See the example below for more.
3. The InitialStateAnalysisWrapper material is somewhat tricky to use in dynamic analysis. Sometimes setting
the displacement to zero appears to be interpreted as an initial displacement in subsequent steps, resulting in
undesirable vibrations.

Initial Stress Material

nDMaterial(’InitStressNDMaterial’, matTag, otherTag, initStress, nDim)

This command is used to construct an Initial Stress material object. The stress-strain behaviour for this material
is defined by another material. Initial Stress Material enables definition of initial stress for the material under
consideration. The strain that corresponds to the initial stress will be calculated from the other material.

matTag (int) integer tag identifying material

otherTag (float) tag of the other material
initStress (float) initial stress
nDim (int) Number of dimensions (e.g. if plane strain nDim=2)

1.4. Model Commands 205

OpenSeesPy Documentation, Release 3.5.1.3

See also:
Notes

Initial Strain Material

nDMaterial(’InitStrainNDMaterial’, matTag, otherTag, initStrain, nDim)

This command is used to construct an Initial Strain material object. The stress-strain behaviour for this material
is defined by another material. Initial Strain Material enables definition of initial strains for the material under
consideration. The stress that corresponds to the initial strain will be calculated from the other material.

matTag (int) integer tag identifying material

otherTag (int) tag of the other material
initStrain (float) initial strain
nDim (float) Number of dimensions

See also:
Notes

UC San Diego soil models

1. PressureIndependMultiYield
2. PressureDependMultiYield
3. PressureDependMultiYield02
4. PressureDependMultiYield03

PressureIndependMultiYield

nDMaterial(’PressureIndependMultiYield’, matTag, nd, rho, refShearModul, refBulkModul, cohesi, peakS-

hearStra, frictionAng=0., refPress=100., pressDependCoe=0., noYieldSurf=20, *yieldSurf )
PressureIndependMultiYield material is an elastic-plastic material in which plasticity exhibits only in the de-
viatoric stress-strain response. The volumetric stress-strain response is linear-elastic and is independent of the
deviatoric response. This material is implemented to simulate monotonic or cyclic response of materials whose
shear behavior is insensitive to the confinement change. Such materials include, for example, organic soils or
clay under fast (undrained) loading conditions.

206 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
nd Number of dimensions, 2 for plane-strain, and 3 for 3D analysis.
(float)
rho Saturated soil mass density.
(float)
(𝐺𝑟 ) Reference low-strain shear modulus, specified at a reference mean effective confining pres-
refShearModul
(float) sure refPress of p’r (see below).
(𝐵𝑟 ) Reference bulk modulus, specified at a reference mean effective confining pressure refPress
refBulkModul
(float) of p’r (see below).
cohesi (𝑐) Apparent cohesion at zero effective confinement.
(float)
(𝛾𝑚𝑎𝑥 ) An octahedral shear strain at which the maximum shear strength is reached, specified at a
peakShearStra
(float) reference mean effective confining pressure refPress of p’r (see below).
(𝑝ℎ𝑖) Friction angle at peak shear strength in degrees, optional (default is 0.0).
frictionAng
(float)
refPress(𝑝′𝑟 ) Reference mean effective confining pressure at which 𝐺𝑟 , 𝐵𝑟 , and 𝛾𝑚𝑎𝑥 are defined, optional
(float) (default is 100. kPa).
(𝑑) A positive constant defining variations of 𝐺 and 𝐵 as a function of instantaneous effective
pressDependCoe
(float) confinement 𝑝′ (default is 0.0)
′
𝐺 = 𝐺𝑟 ( 𝑝𝑝′ )𝑑
𝑟
′
𝐵 = 𝐵𝑟 ( 𝑝𝑝′ )𝑑
𝑟
If 𝜑 = 0, 𝑑 is reset to 0.0.
Number of yield surfaces, optional (must be less than 40, default is 20). The surfaces are generated
noYieldSurf
(float) based on the hyperbolic relation defined in Note 2 below.
yieldSurfInstead of automatic surfaces generation (Note 2), you can define yield surfaces directly based
(list on desired shear modulus reduction curve. To do so, add a minus sign in front of noYieldSurf,
(float)) then provide noYieldSurf pairs of shear strain (r) and modulus ratio (Gs) values. For example, to
define 10 surfaces: yieldSurf = [r1, Gs1, . . . , r10, Gs10]

See also notes

PressureDependMultiYield

nDMaterial(’PressureDependMultiYield’, matTag, nd, rho, refShearModul, refBulkModul, frictionAng,

peakShearStra, refPress, pressDependCoe, PTAng, contrac, *dilat, *liquefac, noYield-
Surf=20.0, *yieldSurf=[], e=0.6, *params=[0.9, 0.02, 0.7, 101.0], c=0.3)
PressureDependMultiYield material is an elastic-plastic material for simulating the essential response character-
istics of pressure sensitive soil materials under general loading conditions. Such characteristics include dilatancy
(shear-induced volume contraction or dilation) and non-flow liquefaction (cyclic mobility), typically exhibited
in sands or silts during monotonic or cyclic loading.

1.4. Model Commands 207

OpenSeesPy Documentation, Release 3.5.1.3

matTag integer tag identifying material

(int)
nd Number of dimensions, 2 for plane-strain, and 3 for 3D analysis.
(float)
rho Saturated soil mass density.
(float)
(𝐺𝑟 ) Reference low-strain shear modulus, specified at a reference mean effective confining pressure
refShearModul
(float) refPress of p’r (see below).
(𝐵𝑟 ) Reference bulk modulus, specified at a reference mean effective confining pressure refPress of
refBulkModul
(float) p’r (see below).
(𝑝ℎ𝑖) Friction angle at peak shear strength in degrees, optional (default is 0.0).
frictionAng
(float)
(𝛾𝑚𝑎𝑥 ) An octahedral shear strain at which the maximum shear strength is reached, specified at a
peakShearStra
(float) reference mean effective confining pressure refPress of p’r (see below).
refPress(𝑝′𝑟 ) Reference mean effective confining pressure at which 𝐺𝑟 , 𝐵𝑟 , and 𝛾𝑚𝑎𝑥 are defined, optional
(float) (default is 100. kPa).
(𝑑) A positive constant defining variations of 𝐺 and 𝐵 as a function of instantaneous effective con-
pressDependCoe
(float) finement 𝑝′ (default is 0.0)
′
𝐺 = 𝐺𝑟 ( 𝑝𝑝′ )𝑑
𝑟
′
𝐵 = 𝐵𝑟 ( 𝑝𝑝′ )𝑑
𝑟
If 𝜑 = 0, 𝑑 is reset to 0.0.
PTAng(𝜑𝑃 𝑇 ) Phase transformation angle, in degrees.
(float)
contrac A non-negative constant defining the rate of shear-induced volume decrease (contraction) or pore
(float) pressure buildup. A larger value corresponds to faster contraction rate.
dilatNon-negative constants defining the rate of shear-induced volume increase (dilation). Larger values
(list correspond to stronger dilation rate. dilat = [dilat1, dilat2].
(float))
liquefacParameters controlling the mechanism of liquefaction-induced perfectly plastic shear strain accu-
(list mulation, i.e., cyclic mobility. Set liquefac[0] = 0 to deactivate this mechanism altogether.
(float))liquefac[0] defines the effective confining pressure (e.g., 10 kPa in SI units or 1.45 psi in English
units) below which the mechanism is in effect. Smaller values should be assigned to denser sands.
Liquefac[1] defines the maximum amount of perfectly plastic shear strain developed at zero ef-
fective confinement during each loading phase. Smaller values should be assigned to denser sands.
Liquefac[2] defines the maximum amount of biased perfectly plastic shear strain 𝛾𝑏 accumulated
at each loading phase under biased shear loading conditions, as 𝛾𝑏 = 𝑙𝑖𝑞𝑢𝑒𝑓 𝑎𝑐[1] × 𝑙𝑖𝑞𝑢𝑒𝑓 𝑎𝑐[2].
Typically, 𝑙𝑖𝑞𝑢𝑒𝑓 𝑎𝑐[2] takes a value between 0.0 and 3.0. Smaller values should be assigned to denser
sands. See the references listed at the end of this chapter for more information.
Number of yield surfaces, optional (must be less than 40, default is 20). The surfaces are generated
noYieldSurf
(float) based on the hyperbolic relation defined in Note 2 below.
If noYieldSurf<0 && >-100, the user defined yield surface is used. You have to provide a
yieldSurf
(list list of 2*(-noYieldSurf), otherwise, the arguments will be messed up. Also don’t provide user
(float))defined yield surface if noYieldSurf>0, it will mess up the argument list too. Instead of automatic
surfaces generation (Note 2), you can define yield surfaces directly based on desired shear modulus
reduction curve. To do so, add a minus sign in front of noYieldSurf, then provide noYieldSurf pairs
of shear strain (r) and modulus ratio (Gs) values. For example, to define 10 surfaces: yieldSurf = [r1,
Gs1, . . . , r10, Gs10]
e Initial void ratio, optional (default is 0.6).
(float)
params params=[cs1, cs2, cs3, pa] defining a straight critical-state line ec in e-p’ space.
(list If cs3=0,
(float))ec = cs1-cs2 log(p’/pa)
else (Li and Wang, JGGE, 124(12)),
ec = cs1-cs2(p’/pa)cs3
208 where pa is atmospheric pressure for normalization (typically 101 kPa in SI Chapter
units, or1.14.65
Developer
psi in
English units). All four constants are optional
c Numerical constant (default value = 0.3 kPa)
(float)
 OpenSeesPy Documentation, Release 3.5.1.3

See also notes

PressureDependMultiYield02

nDMaterial(’PressureDependMultiYield02’, matTag, nd, rho, refShearModul, refBulkModul, frictionAng,

peakShearStra, refPress, pressDependCoe, PTAng, contrac[0], contrac[2], dilat[0], dilat[2],
noYieldSurf=20.0, *yieldSurf=[], contrac[1]=5.0, dilat[1]=3.0, *liquefac=[1.0,0.0],e=0.6,
*params=[0.9, 0.02, 0.7, 101.0], c=0.1)
PressureDependMultiYield02 material is modified from PressureDependMultiYield material, with:
1. additional parameters (contrac[2] and dilat[2]) to account for 𝐾𝜎 effect,
2. a parameter to account for the influence of previous dilation history on subsequent contraction phase
(contrac[1]), and
3. modified logic related to permanent shear strain accumulation (liquefac[0] and liquefac[1]).

matTag (int) integer tag identifying material

nd (float) Number of dimensions, 2 for plane-strain, and 3 for 3D analysis.
rho (float) Saturated soil mass density.
refShearModul(𝐺𝑟 ) Reference low-strain shear modulus, specified at a reference mean effective confining
(float) pressure refPress of p’r (see below).
refBulkModul(𝐵𝑟 ) Reference bulk modulus, specified at a reference mean effective confining pressure
(float) refPress of p’r (see below).
frictionAng (𝑝ℎ𝑖) Friction angle at peak shear strength in degrees, optional (default is 0.0).
(float)
peakShearStra(𝛾𝑚𝑎𝑥 ) An octahedral shear strain at which the maximum shear strength is reached, speci-
(float) fied at a reference mean effective confining pressure refPress of p’r (see below).
refPress (𝑝′𝑟 ) Reference mean effective confining pressure at which 𝐺𝑟 , 𝐵𝑟 , and 𝛾𝑚𝑎𝑥 are defined,
(float) optional (default is 100. kPa).
(𝑑) A positive constant defining variations of 𝐺 and 𝐵 as a function of instantaneous effec-
pressDependCoe
(float) tive confinement 𝑝′ (default is 0.0)
′
𝐺 = 𝐺𝑟 ( 𝑝𝑝′ )𝑑
𝑟
′
𝐵 = 𝐵𝑟 ( 𝑝𝑝′ )𝑑
𝑟
If 𝜑 = 0, 𝑑 is reset to 0.0.
PTAng (𝜑𝑃 𝑇 ) Phase transformation angle, in degrees.
(float)
contrac[2] A non-negative constant reflecting 𝐾𝜎 effect.
(float)
dilat[2] A non-negative constant reflecting 𝐾𝜎 effect.
(float)
contrac[1] A non-negative constant reflecting dilation history on contraction tendency.
(float)
liquefac[0] Damage parameter to define accumulated permanent shear strain as a function of dilation
(float) history. (Redefined and different from PressureDependMultiYield material).
liquefac[1] Damage parameter to define biased accumulation of permanent shear strain as a function of
(float) load reversal history. (Redefined and different from PressureDependMultiYield material).
c (float) Numerical constant (default value = 0.1 kPa)

See also notes

1.4. Model Commands 209

OpenSeesPy Documentation, Release 3.5.1.3

PressureDependMultiYield03

nDMaterial(’PressureDependMultiYield03’, matTag, nd, rho, refShearModul, refBulkModul, frictionAng,

peakShearStra, refPress, pressDependCoe, PTAng, ca, cb, cc, cd, ce, da, db, dc, noYield-
Surf=20.0, *yieldSurf=[], liquefac1=1, liquefac2=0., pa=101, s0=1.73)
The reference for PressureDependMultiYield03 material: Khosravifar, A., Elgamal, A., Lu, J., and Li, J. [2018].
“A 3D model for earthquake-induced liquefaction triggering and post-liquefaction response.” Soil Dynamics
and Earthquake Engineering, 110, 43-52)
PressureDependMultiYield03 is modified from PressureDependMultiYield02 material to comply with the es-
tablished guidelines on the dependence of liquefaction triggering to the number of loading cycles, effective
overburden stress (K𝜎), and static shear stress (K𝛼).
The explanations of parameters
See notes

UC San Diego Saturated Undrained soil

1. FluidSolidPorousMaterial

FluidSolidPorousMaterial

nDMaterial(’FluidSolidPorous’, matTag, nd, soilMatTag, combinedBulkModul, pa=101.0)

FluidSolidPorous material couples the responses of two phases: fluid and solid. The fluid phase response is only
volumetric and linear elastic. The solid phase can be any NDMaterial. This material is developed to simulate
the response of saturated porous media under fully undrained condition.

matTag integer tag identifying material

(int)
nd (float) Number of dimensions, 2 for plane-strain, and 3 for 3D analysis.
soilMatTag The material number for the solid phase material (previously defined).
(int)
Combined undrained bulk modulus 𝐵𝑐 relating changes in pore pressure and volumetric
combinedBulkModul
(float) strain, may be approximated by:
𝐵𝑐 ≈ 𝐵𝑓 /𝑛
where 𝐵𝑓 is the bulk modulus of fluid phase (2.2x106 kPa (or 3.191x105 psi) for water), and
𝑛 the initial porosity.
pa (float) Optional atmospheric pressure for normalization (typically 101 kPa in SI units, or 14.65 psi
in English units)

See also notes

1.4.16 section commands

section(secType, secTag, *secArgs)

This command is used to construct a SectionForceDeformation object, hereto referred to as Section, which
represents force-deformation (or resultant stress-strain) relationships at beam-column and plate sample points.

secType (str) section type

secTag (int) section tag.
secArgs (list) a list of section arguments, must be preceded with *.

210 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

For example,

secType = 'Elastic'
secTag = 1
secArgs = [E, A, Iz]
section(secType, secTag, *secArgs)

The following contain information about available secType:

1. Elastic Section
2. Fiber Section
3. NDFiber Section
4. Wide Flange Section
5. RC Section
6. RCCircular Section
7. Parallel Section
8. Section Aggregator
9. Uniaxial Section
10. Elastic Membrane Plate Section
11. Plate Fiber Section
12. Bidirectional Section
13. Isolator2spring Section
14. LayeredShell

Elastic Section

section(’Elastic’, secTag, E_mod, A, Iz, G_mod=None, alphaY=None)

section(’Elastic’, secTag, E_mod, A, Iz, Iy, G_mod, Jxx, alphaY=None, alphaZ=None)
This command allows the user to construct an ElasticSection. The inclusion of shear deformations is optional.
The dofs for 2D elastic section are [P, Mz], for 3D are [P,Mz,My,T].

secTag (int) unique section tag

E_mod (float) Young’s Modulus
A (float) cross-sectional area of section
Iz (float) second moment of area about the local z-axis
Iy (float) second moment of area about the local y-axis (required for 3D analysis)
G_mod (float) Shear Modulus (optional for 2D analysis, required for 3D analysis)
Jxx (float) torsional moment of inertia of section (required for 3D analysis)
alphaY (float) shear shape factor along the local y-axis (optional)
alphaZ (float) shear shape factor along the local z-axis (optional)

Note: The elastic section can be used in the nonlinear beam column elements, which is useful in the initial stages of
developing a complex model.

1.4. Model Commands 211

OpenSeesPy Documentation, Release 3.5.1.3

Fiber Section

section(’Fiber’, secTag, ’-GJ’, GJ)

This command allows the user to construct a FiberSection object. Each FiberSection object is composed of
Fibers, with each fiber containing a UniaxialMaterial, an area and a location (y,z). The dofs for 2D section are
[P, Mz], for 3D are [P,Mz,My,T].

secTag (int) unique section tag

GJ (float) linear-elastic torsional stiffness assigned to the section

section(’Fiber’, secTag, ’-torsion’, torsionMatTag)

This command allows the user to construct a FiberSection object. Each FiberSection object is composed of
Fibers, with each fiber containing a UniaxialMaterial, an area and a location (y,z). The dofs for 2D section are
[P, Mz], for 3D are [P,Mz,My,T].

secTag (int) unique section tag

torsionMatTag uniaxialMaterial tag assigned to the section for torsional response (can be nonlin-
(int) ear)

Note:
1. The commands below should be called after the section command to generate all the fibers in the section.
2. The patch and layer commands can be used to generate multiple fibers in a single command.

Commands to generate all fibers:

1. Fiber Command
2. Patch Command
3. Layer Command

Fiber Command

fiber(yloc, zloc, A, matTag)

This command allows the user to construct a single fiber and add it to the enclosing FiberSection or NDFiber-
Section.

yloc y coordinate of the fiber in the section (local coordinate system)

(float)
zloc z coordinate of the fiber in the section (local coordinate system)
(float)
A (float) cross-sectional area of fiber
matTag material tag associated with this fiber (UniaxialMaterial tag for a FiberSection and NDMaterial
(int) tag for use in an NDFiberSection).

Patch Command

patch(type, *args)
The patch command is used to generate a number of fibers over a cross-sectional area. Currently there are three
types of cross-section that fibers can be generated: quadrilateral, rectangular and circular.

212 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

patch(’quad’, matTag, numSubdivIJ, numSubdivJK, *crdsI, *crdsJ, *crdsK, *crdsL)

This is the command to generate a quadrilateral shaped patch (the geometry of the patch is defined by four
vertices: I J K L. The coordinates of each of the four vertices is specified in COUNTER CLOCKWISE sequence)

matTag (int)material tag associated with this fiber (UniaxialMaterial tag for a FiberSection and NDMa-
terial tag for use in an NDFiberSection).
numSubdivIJ number of subdivisions (fibers) in the IJ direction.
(int)
numSubdivJK number of subdivisions (fibers) in the JK direction.
(int)
crdsI (list y & z-coordinates of vertex I (local coordinate system)
(float))
crdsJ (list y & z-coordinates of vertex J (local coordinate system)
(float))
crdsK (list y & z-coordinates of vertex K (local coordinate system)
(float))
crdsL (list y & z-coordinates of vertex L (local coordinate system)
(float))

patch(’rect’, matTag, numSubdivY, numSubdivZ, *crdsI, *crdsJ)

This is the command to generate a rectangular patch. The geometry of the patch is defined by coordinates of
vertices: I and J. To ensure positive fiber areas are created, (zJ-zI)/(yJ-yI) should be positive.

matTag (int)material tag associated with this fiber (UniaxialMaterial tag for a FiberSection and NDMa-
terial tag for use in an NDFiberSection).
numSubdivY number of subdivisions (fibers) in local y direction.
(int)
numSubdivZ number of subdivisions (fibers) in local z direction.
(int)
crdsI (list y & z-coordinates of vertex I (local coordinate system)
(float))
crdsJ (list y & z-coordinates of vertex J (local coordinate system)
(float))

patch(’circ’, matTag, numSubdivCirc, numSubdivRad, *center, *rad, *ang)

This is the command to generate a circular shaped patch

matTag (int) material tag associated with this fiber (UniaxialMaterial tag for a FiberSection and ND-
Material tag for use in an NDFiberSection).
numSubdivCirc number of subdivisions (fibers) in the circumferential direction (number of wedges)
(int)
numSubdivRad number of subdivisions (fibers) in the radial direction (number of rings)
(int)
center (list y & z-coordinates of the center of the circle
(float))
rad (list (float)) internal & external radius
ang (list (float)) starting & ending-coordinates angles (degrees)

Layer Command

layer(type, *args)
The layer command is used to generate a number of fibers along a line or a circular arc.

1.4. Model Commands 213

OpenSeesPy Documentation, Release 3.5.1.3

layer(’straight’, matTag, numFiber, areaFiber, *start, *end)

This command is used to construct a straight line of fibers

matTag (int) material tag associated with this fiber (UniaxialMaterial tag for a FiberSection and NDMa-
terial tag for use in an NDFiberSection).
numFiber number of fibers along line
(int)
areaFiber area of each fiber
(float)
start (list y & z-coordinates of first fiber in line (local coordinate system)
(float))
end (list y & z-coordinates of last fiber in line (local coordinate system)
(float))

layer(’circ’, matTag,numFiber,areaFiber,*center,radius,*ang=[0.0,360.0-360/numFiber])
This command is used to construct a line of fibers along a circular arc

matTag (int) material tag associated with this fiber (UniaxialMaterial tag for a FiberSection and NDMa-
terial tag for use in an NDFiberSection).
numFiber number of fibers along line
(int)
areaFiber area of each fiber
(float)
center (list y & z-coordinates of center of circular arc
(float))
radius radius of circlular arc
(float)
ang (list starting and ending angle (optional)
(float))

Fiber Thermal Section

section(’FiberThermal’, secTag, ’-GJ’, GJ=0.0)

This command create a FiberSectionThermal object. The dofs for 2D section are [P, Mz], for 3D are [P,
Mz,My].

Note:
1. The commands below should be called after the section command to generate all the fibers in the section.
2. The patch and layer commands can be used to generate multiple fibers in a single command.

Commands to generate all fibers:

1. Fiber Command
2. Patch Command
3. Layer Command

214 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

NDFiber Section

section(’NDFiber’, secTag)
This command allows the user to construct an NDFiberSection object. Each NDFiberSection object is composed
of NDFibers, with each fiber containing an NDMaterial, an area, and a location (y,z). The NDFiberSection works
for 2D and 3D frame elements and it queries the NDMaterial of each fiber for its axial and shear stresses. In 2D,
stress components 11 and 12 are obtained from each fiber in order to provide stress resultants for axial force,
bending moment, and shear [P, Mz, Vy]. Stress components 11, 12, and 13 lead to all six stress resultants
in 3D [P, Mz, Vy, My, Vz, T].
The NDFiberSection works with any NDMaterial via wrapper classes that perform static condensation of the
stress vector down to the 11, 12, and 13 components, or via specific NDMaterial subclasses that implement the
appropriate fiber stress conditions.

secTag (int) unique section tag

Note:
1. The commands below should be called after the section command to generate all the fibers in the section.
2. The patch and layer commands can be used to generate multiple fibers in a single command.

1. fiber()
2. patch()
3. layer()

Wide Flange Section

section(’WFSection2d’, secTag, matTag, d, tw, bf, tf, Nfw, Nff )

This command allows the user to construct a WFSection2d object, which is an encapsulated fiber representation
of a wide flange steel section appropriate for plane frame analysis.

secTag (int) unique section tag

matTag (int) tag of uniaxialMaterial assigned to each fiber
d (float) section depth
tw (float) web thickness
bf (float) flange width
tf (float) flange thickness
Nfw (float) number of fibers in the web
Nff (float) number of fibers in each flange

Note: The section dimensions d, tw, bf, and tf can be found in the AISC steel manual.

RC Section

section(’RCSection2d’, secTag, coreMatTag, coverMatTag, steelMatTag, d, b, cover_depth, Atop, Abot,

Aside, Nfcore, Nfcover, Nfs)
This command allows the user to construct an RCSection2d object, which is an encapsulated fiber representation

1.4. Model Commands 215

OpenSeesPy Documentation, Release 3.5.1.3

of a rectangular reinforced concrete section with core and confined regions of concrete and single top and bottom
layers of reinforcement appropriate for plane frame analysis.

secTag (int) unique section tag

coreMatTag tag of uniaxialMaterial assigned to each fiber in the core region
(int)
coverMatTag tag of uniaxialMaterial assigned to each fiber in the cover region
(int)
steelMatTag tag of uniaxialMaterial assigned to each reinforcing bar
(int)
d (float) section depth
b (float) section width
cover_depth cover depth (assumed uniform around perimeter)
(float)
Atop (float) area of reinforcing bars in top layer
Abot (float) area of reinforcing bars in bottom layer
Aside (float) area of reinforcing bars on intermediate layers
Nfcore (float) number of fibers through the core depth
Nfcover (float) number of fibers through the cover depth
Nfs (float) number of bars on the top and bottom rows of reinforcement (Nfs-2 bars will be placed
on the side rows)

Note: For more general reinforced concrete section definitions, use the Fiber Section command.

RCCircular Section

section(’RCCircularSection’, secTag, coreMatTag, coverMatTag, steelMatTag, d, cover_depth, Ab,

NringsCore, NringsCover, Nwedges, Nsteel, ’-GJ’, GJ <or ’-torsion’, matTag>)
This command allows the user to construct an RCCircularSection object, which is an encapsulated fiber repre-
sentation of a circular reinforced concrete section with core and confined regions of concrete.

secTag (int) unique section tag

coreMatTag (int) tag of uniaxialMaterial assigned to each fiber in the core region
coverMatTag (int) tag of uniaxialMaterial assigned to each fiber in the cover region
steelMatTag (int) tag of uniaxialMaterial assigned to each reinforcing bar
d (float) section radius
cover_depth (float) cover depth (assumed uniform around perimeter)
Ab (float) area of each reinforcing bar
NringsCore (int) number of fiber rings in the core
NringsCover (int) number of fiber rings in the cover
Nwedges (int) number of fiber wedges for the section
Nsteel (int) number of reinforcing bars
GJ (float) secton torsional stiffness
matTag (int) tag of uniaxialMaterial assigned to section torsion response

Note: One of the -GJ or the -torsion inputs is required

For more general reinforced concrete section definitions, use the Fiber Section command.

216 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Parallel Section

section(’Parallel’, secTag, *SecTags)

Connect sections in parallel.

secTag (int) unique section tag

SecTags (list (int)) tags of of predefined sections.

Section Aggregator

section(’Aggregator’, secTag, *mats, ’-section’, sectionTag)

This command is used to construct a SectionAggregator object which aggregates groups previously-defined
UniaxialMaterial objects into a single section force-deformation model. Each UniaxialMaterial object represents
the section force-deformation response for a particular section degree-of-freedom (dof). There is no interaction
between responses in different dof directions. The aggregation can include one previously defined section.

secTag (int) unique section tag

mats (list) list of tags and dofs of previously-defined Uniax-
ialMaterial objects, mats = [matTag1,dof1,
matTag2,dof2,...]
the force-deformation quantity to be modeled by this
section object. One of the following section dof may
be used:
• 'P' Axial force-deformation
• 'Mz' Moment-curvature about section local z-
axis
• 'Vy' Shear force-deformation along section
local y-axis
• 'My' Moment-curvature about section local y-
axis
• 'Vz' Shear force-deformation along section
local z-axis
• 'T' Torsion Force-Deformation

sectionTag (int) tag of previously-defined Section object to which

the UniaxialMaterial objects are aggregated as addi-
tional force-deformation relationships (optional)

Uniaxial Section

section(’Uniaxial’, secTag, matTag, quantity)

This command is used to construct a UniaxialSection object which uses a previously-defined UniaxialMaterial
object to represent a single section force-deformation response quantity.

1.4. Model Commands 217

OpenSeesPy Documentation, Release 3.5.1.3

secTag (int) unique section tag

matTag (int) tag of uniaxial material
quantity (str) the force-deformation quantity to be modeled by this
section object. One of the following section dof may
be used:
• 'P' Axial force-deformation
• 'Mz' Moment-curvature about section local z-
axis
• 'Vy' Shear force-deformation along section
local y-axis
• 'My' Moment-curvature about section local y-
axis
• 'Vz' Shear force-deformation along section
local z-axis
• 'T' Torsion Force-Deformation

Elastic Membrane Plate Section

section(’ElasticMembranePlateSection’, secTag, E_mod, nu, h, rho, <Ep_modifier>)

This command allows the user to construct an ElasticMembranePlateSection object, which is an isotropic section
appropriate for plate and shell analysis.

secTag (int) unique section tag

E_mod (float) Young’s Modulus
nu (float) Poisson’s Ratio
h (float) depth of section
rho (float) mass density
Ep_modifier factor to modify out-of-plane bending stiffness (optional). E_plate =
(float) Ep_modifier*E_mod

Plate Fiber Section

section(’PlateFiber’, secTag, matTag, h)

This command allows the user to construct a MembranePlateFiberSection object, which is a section that numer-
ically integrates through the plate thickness with “fibers” and is appropriate for plate and shell analysis.

secTag (int) unique section tag

matTag (int) nDMaterial tag to be assigned to each fiber
h (float) plate thickness

Bidirectional Section

section(’Bidirectional’, secTag, E_mod, Fy, Hiso, Hkin, code1=’Vy’, code2=’P’)

This command allows the user to construct a Bidirectional section, which is a stress-resultant plasticity model
of two coupled forces. The yield surface is circular and there is combined isotropic and kinematic hardening.

218 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

secTag (int) unique section tag

E_mod (float) elastic modulus
Fy (float) yield force
Hiso (float) isotropic hardening modulus
Hkin (float) kinematic hardening modulus
code1 (str) section force code for direction 1 (optional)
code2 (str) section force code for direction 2 (optional)
One of the following section code may be used:
• 'P' Axial force-deformation
• 'Mz' Moment-curvature about section local z-
axis
• 'Vy' Shear force-deformation along section
local y-axis
• 'My' Moment-curvature about section local y-
axis
• 'Vz' Shear force-deformation along section
local z-axis
• 'T' Torsion Force-Deformation

Isolator2spring Section

section(’Isolator2spring’, matTag, tol, k1, Fyo, k2o, kvo, hb, PE, Po=0.0)
This command is used to construct an Isolator2spring section object, which represents the buckling behavior of
an elastomeric bearing for two-dimensional analysis in the lateral and vertical plane. An Isolator2spring section
represents the resultant force-deformation behavior of the bearing, and should be used with a zeroLengthSection
element. The bearing should be constrained against rotation.

secTag unique section tag

(int)
tol tolerance for convergence of the element state. Suggested value: E-12 to E-10. OpenSees will
(float) warn if convergence is not achieved, however this usually does not prevent global convergence.
k1 initial stiffness for lateral force-deformation
(float)
Fyo nominal yield strength for lateral force-deformation
(float)
k2o nominal postyield stiffness for lateral force-deformation
(float)
kvo nominal stiffness in the vertical direction
(float)
hb total height of elastomeric bearing
(float)
PE Euler Buckling load for the bearing
(float)
Po axial load at which nominal yield strength is achieved (optional)
(float)

LayeredShell

section(’LayeredShell’, sectionTag, nLayers, *mats)

This command will create the section of the multi-layer shell element, including the multi-dimensional concrete,

1.4. Model Commands 219

OpenSeesPy Documentation, Release 3.5.1.3

reinforcement material and the corresponding thickness.

sectionTag (int) unique tag among sections

nLayers (int) total numbers of layers
mats (list) a list of material tags and thickness, [[mat1,thk1], ..., [mat2,thk2]]

1.4.17 frictionModel commands

frictionModel(frnType, frnTag, *frnArgs)

The frictionModel command is used to construct a friction model object, which specifies the behavior of the
coefficient of friction in terms of the absolute sliding velocity and the pressure on the contact area. The command
has at least one argument, the friction model type.

frnType (str) frictionModel type

frnTag (int) frictionModel tag.
frnArgs (list) a list of frictionModel arguments, must be preceded with *.

For example,

frnType = 'Coulomb'
frnTag = 1
frnArgs = [mu]
frictionModel(frnType, frnTag, *frnArgs)

The following contain information about available frnType:

1. Coulomb
2. Velocity Dependent Friction
3. Velocity and Normal Force Dependent Friction
4. Velocity and Pressure Dependent Friction
5. Multi-Linear Velocity Dependent Friction

Coulomb

frictionModel(’Coulomb’, frnTag, mu)

This command is used to construct a Coulomb friction model object. Coulomb’s Law of Friction states that
kinetic friction is independent of the sliding velocity.

frnTag (int) unique friction model tag

mu (float) coefficient of friction

Velocity Dependent Friction

frictionModel(’VelDependent’, frnTag, muSlow, muFast, transRate)

This command is used to construct a VelDependent friction model object. It is useful for modeling the behavior
of PTFE or PTFE-like materials sliding on a stainless steel surface. For a detailed presentation on the velocity
dependence of such interfaces please refer to Constantinou et al. (1999).

220 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

frnTag (int) unique friction model tag

muSlow (float) coefficient of friction at low velocity
muFast (float) coefficient of friction at high velocity
transRate (float) transition rate from low to high velocity

𝜇 = 𝜇𝑓 𝑎𝑠𝑡 − (𝜇𝑓 𝑎𝑠𝑡 − 𝜇𝑠𝑙𝑜𝑤 ) · 𝑒−𝑡𝑟𝑎𝑛𝑠𝑅𝑎𝑡𝑒 · |𝑣|

REFERENCE:
Constantinou, M.C., Tsopelas, P., Kasalanati, A., and Wolff, E.D. (1999). “Property modification factors for seismic
isolation bearings”. Report MCEER-99-0012, Multidisciplinary Center for Earthquake Engineering Research, State
University of New York.

Velocity and Normal Force Dependent Friction

frictionModel(’VelNormalFrcDep’, frnTag, aSlow, nSlow, aFast, nFast, alpha0, alpha1, alpha2, maxMu-
Fact)
This command is used to construct a VelNormalFrcDep friction model object.

frnTag unique friction model tag

(int)
aSlow constant for coefficient of friction at low velocity
(float)
nSlow exponent for coefficient of friction at low velocity
(float)
aFast constant for coefficient of friction at high velocity
(float)
nFast exponent for coefficient of friction at high velocity
(float)
alpha0 constant rate parameter coefficient
(float)
alpha1 linear rate parameter coefficient
(float)
alpha2 quadratic rate parameter coefficient
(float)
factor for determining the maximum coefficient of friction. This value prevents the friction coef-
maxMuFact
(float) ficient from exceeding an unrealistic maximum value when the normal force becomes very small.
The maximum friction coefficient is determined from 𝜇Fast, for example 𝜇 ≤ 𝑚𝑎𝑥𝑀 𝑢𝐹 𝑎𝑐*𝐹 𝑎𝑠𝑡.

Velocity and Pressure Dependent Friction

frictionModel(’VelPressureDep’, frnTag, muSlow, muFast0, A, deltaMu, alpha, transRate)

This command is used to construct a VelPressureDep friction model object.

1.4. Model Commands 221

OpenSeesPy Documentation, Release 3.5.1.3

frnTag (int) unique friction model tag

muSlow (float) coefficient of friction at low velocity
muFast0 (float) initial coefficient of friction at high velocity
A (float) nominal contact area
deltaMu (float) pressure parameter calibrated from experimental data
alpha (float) pressure parameter calibrated from experimental data
transRate (float) transition rate from low to high velocity

Multi-Linear Velocity Dependent Friction

frictionModel(’VelDepMultiLinear’, frnTag, ’-vel’, *velPoints, ’-frn’, *frnPoints)

This command is used to construct a VelDepMultiLinear friction model object. The friction-velocity relationship
is given by a multi-linear curve that is define by a set of points. The slope given by the last two specified points
on the positive velocity axis is extrapolated to infinite positive velocities. Velocity and friction points need to be
equal or larger than zero (no negative values should be defined). The number of provided velocity points needs
to be equal to the number of provided friction points.

frnTag (int) unique friction model tag

velPoints (list (float)) list of velocity points along friction-velocity curve
frnPoints (list (float)) list of friction points along friction-velocity curve

1.4.18 geomTransf commands

geomTransf(transfType, transfTag, *transfArgs)

The geometric-transformation command is used to construct a coordinate-transformation (CrdTransf) object,
which transforms beam element stiffness and resisting force from the basic system to the global-coordinate
system. The command has at least one argument, the transformation type.

transfType (str) geomTransf type

transfTag (int) geomTransf tag.
transfArgs (list) a list of geomTransf arguments, must be preceded with *.

For example,

transfType = 'Linear'
transfTag = 1
transfArgs = []
geomTransf(transfType, transfTag, *transfArgs)

The following contain information about available transfType:

1. Linear Transformation
2. PDelta Transformation
3. Corotational Transformation

Linear Transformation

geomTransf(’Linear’, transfTag, ’-jntOffset’, *dI, *dJ)

222 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

geomTransf(’Linear’, transfTag, *vecxz, ’-jntOffset’, *dI, *dJ)

This command is used to construct a linear coordinate transformation (LinearCrdTransf) object, which performs
a linear geometric transformation of beam stiffness and resisting force from the basic system to the global-
coordinate system.

transfTaginteger tag identifying transformation

(int)
vecxz X, Y, and Z components of vecxz, the vector used to define the local x-z plane of the local-coordinate
(list system. The local y-axis is defined by taking the cross product of the vecxz vector and the x-axis.
(float)) These components are specified in the global-coordinate system X,Y,Z and define a vector that is in
a plane parallel to the x-z plane of the local-coordinate system. These items need to be specified for
the three-dimensional problem.
dI joint offset values – offsets specified with respect to the global coordinate system for element-end
(list node i (the number of arguments depends on the dimensions of the current model).
(float))
dJ joint offset values – offsets specified with respect to the global coordinate system for element-end
(list node j (the number of arguments depends on the dimensions of the current model).
(float))

PDelta Transformation

geomTransf(’PDelta’, transfTag, ’-jntOffset’, *dI, *dJ)

geomTransf(’PDelta’, transfTag, *vecxz, ’-jntOffset’, *dI, *dJ)
This command is used to construct the P-Delta Coordinate Transformation (PDeltaCrdTransf) object, which
performs a linear geometric transformation of beam stiffness and resisting force from the basic system to the
global coordinate system, considering second-order P-Delta effects.

transfTaginteger tag identifying transformation

(int)
vecxz X, Y, and Z components of vecxz, the vector used to define the local x-z plane of the local-coordinate
(list system. The local y-axis is defined by taking the cross product of the vecxz vector and the x-axis.
(float)) These components are specified in the global-coordinate system X,Y,Z and define a vector that is in
a plane parallel to the x-z plane of the local-coordinate system. These items need to be specified for
the three-dimensional problem.
dI joint offset values – offsets specified with respect to the global coordinate system for element-end
(list node i (the number of arguments depends on the dimensions of the current model).
(float))
dJ joint offset values – offsets specified with respect to the global coordinate system for element-end
(list node j (the number of arguments depends on the dimensions of the current model).
(float))

Note: P LARGE Delta effects do not include P small delta effects.

Corotational Transformation

geomTransf(’Corotational’, transfTag, ’-jntOffset’, *dI, *dJ)

geomTransf(’Corotational’, transfTag, *vecxz)
This command is used to construct the Corotational Coordinate Transformation (CorotCrdTransf) object. Coro-
tational transformation can be used in large displacement-small strain problems.

1.4. Model Commands 223

OpenSeesPy Documentation, Release 3.5.1.3

transfTaginteger tag identifying transformation

(int)
vecxz X, Y, and Z components of vecxz, the vector used to define the local x-z plane of the local-coordinate
(list system. The local y-axis is defined by taking the cross product of the vecxz vector and the x-axis.
(float)) These components are specified in the global-coordinate system X,Y,Z and define a vector that is in
a plane parallel to the x-z plane of the local-coordinate system. These items need to be specified for
the three-dimensional problem.
dI joint offset values – offsets specified with respect to the global coordinate system for element-end
(list node i (the number of arguments depends on the dimensions of the current model).
(float))
dJ joint offset values – offsets specified with respect to the global coordinate system for element-end
(list node j (the number of arguments depends on the dimensions of the current model).
(float))

Note: Currently the transformation does not deal with element loads and will ignore any that are applied to the
element.

1.5 Analysis Commands

In OpenSees, an analysis is an object which is composed by the aggregation of component objects. It is the component
objects which define the type of analysis that is performed on the model. The component classes, as shown in the
figure below, consist of the following:
1. ConstraintHandler – determines how the constraint equations are enforced in the analysis – how it handles the
boundary conditions/imposed displacements
2. DOF_Numberer – determines the mapping between equation numbers and degrees-of-freedom
3. Integrator – determines the predictive step for time t+dt
4. SolutionAlgorithm – determines the sequence of steps taken to solve the non-linear equation at the current time
step
5. SystemOfEqn/Solver – within the solution algorithm, it specifies how to store and solve the system of equations
in the analysis
6. Convergence Test – determines when convergence has been achieved.
Analysis commands
1. constraints commands
2. numberer commands
3. system commands
4. test commands
5. algorithm commands
6. integrator commands
7. analysis command
8. eigen command
9. analyze command

224 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

10. modalProperties Command

11. responseSpectrumAnalysis Command

1.5.1 constraints commands

constraints(constraintType, *constraintArgs)
This command is used to construct the ConstraintHandler object. The ConstraintHandler object determines how
the constraint equations are enforced in the analysis. Constraint equations enforce a specified value for a DOF,
or a relationship between DOFs.

constraintType (str) constraints type

constraintArgs (list) a list of constraints arguments

The following contain information about available constraintType:

1. Plain Constraints
2. Lagrange Multipliers
3. Penalty Method
4. Transformation Method

Plain Constraints

constraints(’Plain’)
This command is used to construct a Plain constraint handler. A plain constraint handler can only enforce
homogeneous single point constraints (fix command) and multi-point constraints constructed where the con-
straint matrix is equal to the identity (equalDOF command). The following is the command to construct a plain
constraint handler:

Note: As mentioned, this constraint handler can only enforce homogeneous single point constraints (fix command)
and multi-pont constraints where the constraint matrix is equal to the identity (equalDOF command).

Lagrange Multipliers

constraints(’Lagrange’, alphaS=1.0, alphaM=1.0)

This command is used to construct a LagrangeMultiplier constraint handler, which enforces the constraints by
introducing Lagrange multiplies to the system of equation. The following is the command to construct a plain
constraint handler:

alphaS (float) 𝛼𝑆 factor on single points.

alphaM (float) 𝛼𝑀 factor on multi-points.

Note: The Lagrange multiplier method introduces new unknowns to the system of equations. The diagonal part of the
system corresponding to these new unknowns is 0.0. This ensure that the system IS NOT symmetric positive definite.

1.5. Analysis Commands 225

OpenSeesPy Documentation, Release 3.5.1.3

Penalty Method

constraints(’Penalty’, alphaS=1.0, alphaM=1.0)

This command is used to construct a Penalty constraint handler, which enforces the constraints using the penalty
method. The following is the command to construct a penalty constraint handler:

alphaS (float) 𝛼𝑆 factor on single points.

alphaM (float) 𝛼𝑀 factor on multi-points.

Note: The degree to which the constraints are enforced is dependent on the penalty values chosen. Problems can
arise if these values are too small (constraint not enforced strongly enough) or too large (problems associated with
conditioning of the system of equations).

Transformation Method

constraints(’Transformation’)
This command is used to construct a transformation constraint handler, which enforces the constraints using the
transformation method. The following is the command to construct a transformation constraint handler

Note:
• The single-point constraints when using the transformation method are done directly. The matrix equation is
not manipulated to enforce them, rather the trial displacements are set directly at the nodes at the start of each
analysis step.
• Great care must be taken when multiple constraints are being enforced as the transformation method does not
follow constraints:
1. If a node is fixed, constrain it with the fix command and not equalDOF or other type of constraint.
2. If multiple nodes are constrained, make sure that the retained node is not constrained in any other con-
straint.
And remember if a node is constrained to multiple nodes in your model it probably means you have messed up.

1.5.2 numberer commands

numberer(numbererType, *numbererArgs)
This command is used to construct the DOF_Numberer object. The DOF_Numberer object determines the
mapping between equation numbers and degrees-of-freedom – how degrees-of-freedom are numbered.

numbererType (str) numberer type

numbererArgs (list) a list of numberer arguments

The following contain information about available numbererType:

1. Plain Numberer
2. RCM Numberer
3. AMD Numberer

226 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

4. Parallel Plain Numberer

5. Parallel RCM Numberer

Plain Numberer

numberer(’Plain’)
This command is used to construct a Plain degree-of-freedom numbering object to provide the mapping between
the degrees-of-freedom at the nodes and the equation numbers. A Plain numberer just takes whatever order the
domain gives it nodes and numbers them, this ordering is both dependent on node numbering and size of the
model.

Note: For very small problems and for the sparse matrix solvers which provide their own numbering scheme, order
is not really important so plain numberer is just fine. For large models and analysis using solver types other than the
sparse solvers, the order will have a major impact on performance of the solver and the plain handler is a poor choice.

RCM Numberer

numberer(’RCM’)
This command is used to construct an RCM degree-of-freedom numbering object to provide the mapping be-
tween the degrees-of-freedom at the nodes and the equation numbers. An RCM numberer uses the reverse
Cuthill-McKee scheme to order the matrix equations.

AMD Numberer

numberer(’AMD’)
This command is used to construct an AMD degree-of-freedom numbering object to provide the mapping be-
tween the degrees-of-freedom at the nodes and the equation numbers. An AMD numberer uses the approximate
minimum degree scheme to order the matrix equations.

Parallel Plain Numberer

numberer(’ParallelPlain’)
This command is used to construct a parallel version of Plain degree-of-freedom numbering object to provide
the mapping between the degrees-of-freedom at the nodes and the equation numbers. A Plain numberer just
takes whatever order the domain gives it nodes and numbers them, this ordering is both dependent on node
numbering and size of the model.
Use this command only for parallel model.

Warning: Don’t use this command if model is not parallel, for example, parametric study.

Parallel RCM Numberer

numberer(’ParallelRCM’)
This command is used to construct a parallel version of RCM degree-of-freedom numbering object to provide
the mapping between the degrees-of-freedom at the nodes and the equation numbers. A Plain numberer just
takes whatever order the domain gives it nodes and numbers them, this ordering is both dependent on node
numbering and size of the model.

1.5. Analysis Commands 227

OpenSeesPy Documentation, Release 3.5.1.3

Use this command only for parallel model.

Warning: Don’t use this command if model is not parallel, for example, parametric study.

1.5.3 system commands

system(systemType, *systemArgs)
This command is used to construct the LinearSOE and LinearSolver objects to store and solve the system of
equations in the analysis.

systemType (str) system type

systemArgs (list) a list of system arguments

The following contain information about available systemType:

1. BandGeneral SOE
2. BandSPD SOE
3. ProfileSPD SOE
4. SuperLU SOE
5. UmfPack SOE
6. FullGeneral SOE
7. SparseSYM SOE
8. PFEM SOE
9. MUMPS Solver

BandGeneral SOE

system(’BandGen’)
This command is used to construct a BandGeneralSOE linear system of equation object. As the name implies,
this class is used for matrix systems which have a banded profile. The matrix is stored as shown below in a
1dimensional array of size equal to the bandwidth times the number of unknowns. When a solution is required,
the Lapack routines DGBSV and SGBTRS are used.

BandSPD SOE

system(’BandSPD’)
This command is used to construct a BandSPDSOE linear system of equation object. As the name implies, this
class is used for symmetric positive definite matrix systems which have a banded profile. The matrix is stored as
shown below in a 1 dimensional array of size equal to the (bandwidth/2) times the number of unknowns. When
a solution is required, the Lapack routines DPBSV and DPBTRS are used.

228 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

ProfileSPD SOE

system(’ProfileSPD’)
This command is used to construct a profileSPDSOE linear system of equation object. As the name implies,
this class is used for symmetric positive definite matrix systems. The matrix is stored as shown below in a
1 dimensional array with only those values below the first non-zero row in any column being stored. This is
sometimes also referred to as a skyline storage scheme.

SuperLU SOE

system(’SuperLU’)
This command is used to construct a SparseGEN linear system of equation object. As the name implies, this
class is used for sparse matrix systems. The solution of the sparse matrix is carried out using SuperLU.

UmfPack SOE

system(’UmfPack’)
This command is used to construct a sparse system of equations which uses the UmfPack solver.

FullGeneral SOE

system(’FullGeneral’)
This command is used to construct a Full General linear system of equation object. As the name implies, the
class utilizes NO space saving techniques to cut down on the amount of memory used. If the matrix is of size,
nxn, then storage for an nxn array is sought from memory when the program runs. When a solution is required,
the Lapack routines DGESV and DGETRS are used.

Note: This type of system should almost never be used! This is because it requires a lot more memory than every
other solver and takes more time in the actal solving operation than any other solver. It is required if the user is
interested in looking at the global system matrix.

SparseSYM SOE

system(’SparseSYM’)
This command is used to construct a sparse symmetric system of equations which uses a row-oriented solution
method in the solution phase.

MUMPS Solver

system(’Mumps’, ’-ICNTL14’, icntl14=20.0, ’-ICNTL7’, icntl7=7)

Create a system of equations using the Mumps solver

1.5. Analysis Commands 229

OpenSeesPy Documentation, Release 3.5.1.3

icntl14 controls the percentage increase in the estimated

working space (optional)
icntl7 computes a symmetric permutation (ordering) to de-
termine the pivot order to be used for the factoriza-
tion in case of sequential analysis (optional)
• 0: AMD
• 1: set by user
• 2: AMF
• 3: SCOTCH
• 4: PORD
• 5: Metis
• 6: AMD with QADM
• 7: automatic

Use this command only for parallel model.

Warning: Don’t use this command if model is not parallel, for example, parametric study.

1.5.4 test commands

test(testType, *testArgs)
This command is used to construct the LinearSOE and LinearSolver objects to store and solve the test of equa-
tions in the analysis.

testType (str) test type

testArgs (list) a list of test arguments

The following contain information about available testType:

1. NormUnbalance
2. NormDispIncr
3. energyIncr
4. RelativeNormUnbalance
5. RelativeNormDispIncr
6. RelativeTotalNormDispIncr
7. RelativeEnergyIncr
8. FixedNumIter
9. NormDispAndUnbalance
10. NormDispOrUnbalance
11. PFEM test

230 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

NormUnbalance

test(’NormUnbalance’, tol, iter, pFlag=0, nType=2, maxIncr=maxIncr)

Create a NormUnbalance test, which uses the norm of the right hand side of the matrix equation to determine if
convergence has been reached.

tol (float) Tolerance criteria used to check for convergence.

iter (int) Max number of iterations to check
pFlag (int) Print flag (optional):
• 0 print nothing.
• 1 print information on norms each time
test() is invoked.
• 2 print information on norms and number of it-
erations at end of successful test.
• 4 at each step it will print the norms and also
the ∆𝑈 and 𝑅(𝑈 ) vectors.
• 5 if it fails to converge at end of numIter it
will print an error message but return a suc-
cessfull test.

nType (int) Type of norm, (0 = max-norm, 1 = 1-norm, 2 = 2-

norm). (optional)
maxIncr (int) Maximum times of error increasing. (optional)

When using the Penalty method additional large forces to enforce the penalty functions exist on the right hand
side, making convergence using this test usually impossible (even though solution might have converged).

NormDispIncr

test(’NormDispIncr’, tol, iter, pFlag=0, nType=2)

Create a NormDispIncr test, which uses the norm of the left hand side solution vector of the matrix equation to
determine if convergence has been reached.

tol (float) Tolerance criteria used to check for convergence.

iter (int) Max number of iterations to check
pFlag (int) Print flag (optional):
• 0 print nothing.
• 1 print information on norms each time
test() is invoked.
• 2 print information on norms and number of it-
erations at end of successful test.
• 4 at each step it will print the norms and also
the ∆𝑈 and 𝑅(𝑈 ) vectors.
• 5 if it fails to converge at end of numIter it
will print an error message but return a suc-
cessfull test.

nType (int) Type of norm, (0 = max-norm, 1 = 1-norm, 2 = 2-

norm). (optional)

When using the Lagrange method to enforce the constraints, the Lagrange multipliers appear in the solution
vector.

1.5. Analysis Commands 231

OpenSeesPy Documentation, Release 3.5.1.3

energyIncr

test(’EnergyIncr’, tol, iter, pFlag=0, nType=2)

Create a EnergyIncr test, which uses the dot product of the solution vector and norm of the right hand side of
the matrix equation to determine if convergence has been reached.

tol (float) Tolerance criteria used to check for convergence.

iter (int) Max number of iterations to check
pFlag (int) Print flag (optional):
• 0 print nothing.
• 1 print information on norms each time
test() is invoked.
• 2 print information on norms and number of it-
erations at end of successful test.
• 4 at each step it will print the norms and also
the ∆𝑈 and 𝑅(𝑈 ) vectors.
• 5 if it fails to converge at end of numIter it
will print an error message but return a suc-
cessfull test.

nType (int) Type of norm, (0 = max-norm, 1 = 1-norm, 2 = 2-

norm). (optional)

• When using the Penalty method additional large forces to enforce the penalty functions exist on the right
hand side, making convergence using this test usually impossible (even though solution might have con-
verged).
• When using the Lagrange method to enforce the constraints, the Lagrange multipliers appear in the solution
vector.

RelativeNormUnbalance

test(’RelativeNormUnbalance’, tol, iter, pFlag=0, nType=2)

Create a RelativeNormUnbalance test, which uses the relative norm of the right hand side of the matrix equation
to determine if convergence has been reached.

tol (float) Tolerance criteria used to check for convergence.

iter (int) Max number of iterations to check
pFlag (int) Print flag (optional):
• 0 print nothing.
• 1 print information on norms each time
test() is invoked.
• 2 print information on norms and number of it-
erations at end of successful test.
• 4 at each step it will print the norms and also
the ∆𝑈 and 𝑅(𝑈 ) vectors.
• 5 if it fails to converge at end of numIter it
will print an error message but return a suc-
cessfull test.

nType (int) Type of norm, (0 = max-norm, 1 = 1-norm, 2 = 2-

norm). (optional)

232 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

• When using the Penalty method additional large forces to enforce the penalty functions exist on the right
hand side, making convergence using this test usually impossible (even though solution might have con-
verged).

RelativeNormDispIncr

test(’RelativeNormDispIncr’, tol, iter, pFlag=0, nType=2)

Create a RelativeNormDispIncr test, which uses the relative of the solution vector of the matrix equation to
determine if convergence has been reached.

tol (float) Tolerance criteria used to check for convergence.

iter (int) Max number of iterations to check
pFlag (int) Print flag (optional):
• 0 print nothing.
• 1 print information on norms each time
test() is invoked.
• 2 print information on norms and number of it-
erations at end of successful test.
• 4 at each step it will print the norms and also
the ∆𝑈 and 𝑅(𝑈 ) vectors.
• 5 if it fails to converge at end of numIter it
will print an error message but return a suc-
cessfull test.

nType (int) Type of norm, (0 = max-norm, 1 = 1-norm, 2 = 2-

norm). (optional)

RelativeTotalNormDispIncr

test(’RelativeTotalNormDispIncr’, tol, iter, pFlag=0, nType=2)

Create a RelativeTotalNormDispIncr test, which uses the ratio of the current norm to the total norm (the sum of
all the norms since last convergence) of the solution vector.

tol (float) Tolerance criteria used to check for convergence.

iter (int) Max number of iterations to check
pFlag (int) Print flag (optional):
• 0 print nothing.
• 1 print information on norms each time
test() is invoked.
• 2 print information on norms and number of it-
erations at end of successful test.
• 4 at each step it will print the norms and also
the ∆𝑈 and 𝑅(𝑈 ) vectors.
• 5 if it fails to converge at end of numIter it
will print an error message but return a suc-
cessfull test.

nType (int) Type of norm, (0 = max-norm, 1 = 1-norm, 2 = 2-

norm). (optional)

1.5. Analysis Commands 233

OpenSeesPy Documentation, Release 3.5.1.3

RelativeEnergyIncr

test(’RelativeEnergyIncr’, tol, iter, pFlag=0, nType=2)

Create a RelativeEnergyIncr test, which uses the relative dot product of the solution vector and norm of the right
hand side of the matrix equation to determine if convergence has been reached.

tol (float) Tolerance criteria used to check for convergence.

iter (int) Max number of iterations to check
pFlag (int) Print flag (optional):
• 0 print nothing.
• 1 print information on norms each time
test() is invoked.
• 2 print information on norms and number of it-
erations at end of successful test.
• 4 at each step it will print the norms and also
the ∆𝑈 and 𝑅(𝑈 ) vectors.
• 5 if it fails to converge at end of numIter it
will print an error message but return a suc-
cessfull test.

nType (int) Type of norm, (0 = max-norm, 1 = 1-norm, 2 = 2-

norm). (optional)

FixedNumIter

test(’FixedNumIter’, iter, pFlag=0, nType=2)

Create a FixedNumIter test, that performs a fixed number of iterations without testing for convergence.

tol (float) Tolerance criteria used to check for convergence.

iter (int) Max number of iterations to check
pFlag (int) Print flag (optional):
• 0 print nothing.
• 1 print information on norms each time
test() is invoked.
• 2 print information on norms and number of it-
erations at end of successful test.
• 4 at each step it will print the norms and also
the ∆𝑈 and 𝑅(𝑈 ) vectors.
• 5 if it fails to converge at end of numIter it
will print an error message but return a suc-
cessfull test.

nType (int) Type of norm, (0 = max-norm, 1 = 1-norm, 2 = 2-

norm). (optional)

NormDispAndUnbalance

test(’NormDispAndUnbalance’, tolIncr, tolR, iter, pFlag=0, nType=2, maxincr=-1)

Create a NormDispAndUnbalance test, which check if both 'NormUnbalance' and 'NormDispIncr' are
converged.

234 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

tolIncr (float) Tolerance for left hand solution increments

tolIncr (float) Tolerance for right hand residual
iter (int) Max number of iterations to check
pFlag (int) Print flag (optional):
• 0 print nothing.
• 1 print information on norms each time
test() is invoked.
• 2 print information on norms and number of it-
erations at end of successful test.
• 4 at each step it will print the norms and also
the ∆𝑈 and 𝑅(𝑈 ) vectors.
• 5 if it fails to converge at end of numIter it
will print an error message but return a suc-
cessfull test.

nType (int) Type of norm, (0 = max-norm, 1 = 1-norm, 2 = 2-

norm). (optional)
maxincr (int) Maximum times of error increasing. (optional)

NormDispOrUnbalance

test(’NormDispOrUnbalance’, tolIncr, tolR, iter, pFlag=0, nType=2, maxincr=-1)

Create a NormDispOrUnbalance test, which check if both 'NormUnbalance' and 'normDispIncr' are
converged.

tolIncr (float) Tolerance for left hand solution increments

tolIncr (float) Tolerance for right hand residual
iter (int) Max number of iterations to check
pFlag (int) Print flag (optional):
• 0 print nothing.
• 1 print information on norms each time
test() is invoked.
• 2 print information on norms and number of it-
erations at end of successful test.
• 4 at each step it will print the norms and also
the ∆𝑈 and 𝑅(𝑈 ) vectors.
• 5 if it fails to converge at end of numIter it
will print an error message but return a suc-
cessfull test.

nType (int) Type of norm, (0 = max-norm, 1 = 1-norm, 2 = 2-

norm). (optional)
maxincr (int) Maximum times of error increasing. (optional)

1.5.5 algorithm commands

algorithm(algoType, *algoArgs)
This command is used to construct a SolutionAlgorithm object, which determines the sequence of steps taken
to solve the non-linear equation.

1.5. Analysis Commands 235

OpenSeesPy Documentation, Release 3.5.1.3

algoType (str) algorithm type

algoArgs (list) a list of algorithm arguments

The following contain information about available algoType:

1. Linear Algorithm
2. Newton Algorithm
3. Newton with Line Search
4. Modified Newton Algorithm
5. Krylov-Newton Algorithm
6. SecantNewton Algorithm
7. RaphsonNewton Algorithm
8. PeriodicNewton Algorithm
9. BFGS Algorithm
10. Broyden Algorithm

Linear Algorithm

algorithm(’Linear’, secant=False, initial=False, factorOnce=False)

Create a Linear algorithm which takes one iteration to solve the system of equations.

secant (bool) Flag to indicate to use secant stiffness. (optional)

initial (bool) Flag to indicate to use initial stiffness. (optional)
factorOnce (bool) Flag to indicate to only set up and factor matrix once. (optional)

Note: As the tangent matrix typically will not change during the analysis in case of an elastic system it is highly
advantageous to use the -factorOnce option. Do not use this option if you have a nonlinear system and you want the
tangent used to be actual tangent at time of the analysis step.

Newton Algorithm

algorithm(’Newton’, secant=False, initial=False, initialThenCurrent=False)

Create a Newton-Raphson algorithm. The Newton-Raphson method is the most widely used and most robust
method for solving nonlinear algebraic equations.

secant (bool) Flag to indicate to use secant stiffness. (optional)

initial (bool) Flag to indicate to use initial stiffness.(optional)
initialThenCurrent Flag to indicate to use initial stiffness on first step, then use current stiffness for
(bool) subsequent steps. (optional)

236 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Newton with Line Search

algorithm(’NewtonLineSearch’, Bisection=False, Secant=False, RegulaFalsi=False, InitialInterpo-

lated=False, tol=0.8, maxIter=10, minEta=0.1, maxEta=10.0)
Create a NewtonLineSearch algorithm. Introduces line search to the Newton algorithm to solve the nonlinear
residual equation.

Bisection (bool) Flag to use Bisection line search. (optional)

Secant (bool) Flag to use Secant line search. (optional)
RegulaFalsi (bool) Flag to use RegulaFalsi line search. (optional)
InitialInterpolated (bool) Flag to use InitialInterpolated line search.(optional)
tol (float) Tolerance for search. (optional)
maxIter (float) Max num of iterations to try. (optional)
minEta (float) Min 𝜂 value. (optional)
maxEta (float) Max 𝜂 value. (optional)

Modified Newton Algorithm

algorithm(’ModifiedNewton’, secant=False, initial=False)

Create a ModifiedNewton algorithm. The difference to Newton is that the tangent at the initial guess is used in
the iterations, instead of the current tangent.

secant (bool) Flag to indicate to use secant stiffness. (optional)

initial (bool) Flag to indicate to use initial stiffness.(optional)

Krylov-Newton Algorithm

algorithm(’KrylovNewton’, iterate=’current’, increment=’current’, maxDim=3)

Create a KrylovNewton algorithm which uses a Krylov subspace accelerator to accelerate the convergence of
the ModifiedNewton.

iterate (str) Tangent to iterate on, 'current', 'initial', 'noTangent' (optional)

increment Tangent to increment on, 'current', 'initial', 'noTangent' (optional)
(str)
maxDim (int) Max number of iterations until the tangent is reformed and the acceleration restarts.
(optional)

SecantNewton Algorithm

algorithm(’SecantNewton’, iterate=’current’, increment=’current’, maxDim=3)

Create a SecantNewton algorithm which uses the two-term update to accelerate the convergence of the Modi-
fiedNewton.
The default “cut-out” values recommended by Crisfield (R1=3.5, R2=0.3) are used.

iterate (str) Tangent to iterate on, 'current', 'initial', 'noTangent' (optional)

increment Tangent to increment on, 'current', 'initial', 'noTangent' (optional)
(str)
maxDim (int) Max number of iterations until the tangent is reformed and the acceleration restarts.
(optional)

1.5. Analysis Commands 237

OpenSeesPy Documentation, Release 3.5.1.3

RaphsonNewton Algorithm

algorithm(’RaphsonNewton’, iterate=’current’, increment=’current’)

Create a RaphsonNewton algorithm which uses Raphson accelerator.

iterate (str) Tangent to iterate on, 'current', 'initial', 'noTangent' (optional)

increment (str) Tangent to increment on, 'current', 'initial', 'noTangent' (optional)

PeriodicNewton Algorithm

algorithm(’PeriodicNewton’, iterate=’current’, increment=’current’, maxDim=3)

Create a PeriodicNewton algorithm using periodic accelerator.

iterate (str) Tangent to iterate on, 'current', 'initial', 'noTangent' (optional)

increment Tangent to increment on, 'current', 'initial', 'noTangent' (optional)
(str)
maxDim (int) Max number of iterations until the tangent is reformed and the acceleration restarts.
(optional)

BFGS Algorithm

algorithm(’BFGS’, secant=False, initial=False, count=10)

Create a BFGS algorithm. The BFGS method is one of the most effective matrix-update or quasi Newton
methods for iteration on a nonlinear system of equations. The method computes new search directions at each
iteration step based on the initial jacobian, and subsequent trial solutions. The unlike regular Newton does not
require the tangent matrix be reformulated and refactored at every iteration, however unlike ModifiedNewton it
does not rely on the tangent matrix from a previous iteration.

secant (bool) Flag to indicate to use secant stiffness. (optional)

initial (bool) Flag to indicate to use initial stiffness.(optional)
count (int) Number of iterations. (optional)

Broyden Algorithm

algorithm(’Broyden’, secant=False, initial=False, count=10)

Create a Broyden algorithm for general unsymmetric systems which performs successive rank-one updates of
the tangent at the first iteration of the current time step.

secant (bool) Flag to indicate to use secant stiffness. (optional)

initial (bool) Flag to indicate to use initial stiffness.(optional)
count (int) Number of iterations. (optional)

1.5.6 integrator commands

integrator(intType, *intArgs)
This command is used to construct the Integrator object. The Integrator object determines the meaning of the
terms in the system of equation object Ax=B.
The Integrator object is used for the following:

238 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

• determine the predictive step for time t+dt

• specify the tangent matrix and residual vector at any iteration
• determine the corrective step based on the displacement increment dU

intType (str) integrator type

intArgs (list) a list of integrator arguments

The following contain information about available intType:

Static integrator objects

1. LoadControl
2. DisplacementControl
3. Parallel DisplacementControl
4. Minimum Unbalanced Displacement Norm
5. Arc-Length Control

LoadControl

integrator(’LoadControl’, incr, numIter=1, minIncr=incr, maxIncr=incr)

Create a OpenSees LoadControl integrator object.

incr (float) Load factor increment 𝜆.

numIter (int) Number of iterations the user would like to occur in the solution algorithm. (optional)
minIncr (float) Min stepsize the user will allow 𝜆𝑚𝑖𝑛 . (optional)
maxIncr (float) Max stepsize the user will allow 𝜆𝑚𝑎𝑥 . (optional)

1. The change in applied loads that this causes depends on the active load pattern (those load pattern not
set constant) and the loads in the load pattern. If the only active load acting on the Domain are in load
pattern with a Linear time series with a factor of 1.0, this integrator is the same as the classical load control
method.
2. The optional arguments are supplied to speed up the step size in cases where convergence is too fast and
slow down the step size in cases where convergence is too slow.

DisplacementControl

integrator(’DisplacementControl’, nodeTag, dof, incr, numIter=1, dUmin=incr, dUmax=incr)

Create a DisplacementControl integrator. In an analysis step with Displacement Control we seek to determine
the time step that will result in a displacement increment for a particular degree-of-freedom at a node to be a
prescribed value.

nodeTag (int) tag of node whose response controls solution

dof (int) Degree of freedom at the node, 1 through ndf.
incr (float) First displacement increment ∆𝑈𝑑𝑜𝑓 .
numIter (int) Number of iterations the user would like to occur in the solution algorithm. (optional)
minIncr (float) Min stepsize the user will allow ∆𝑈𝑚𝑖𝑛 . (optional)
maxIncr (float) Max stepsize the user will allow ∆𝑈𝑚𝑎𝑥 . (optional)

1.5. Analysis Commands 239

OpenSeesPy Documentation, Release 3.5.1.3

Parallel DisplacementControl

integrator(’ParallelDisplacementControl’, nodeTag, dof, incr, numIter=1, dUmin=incr, dUmax=incr)

Create a Parallel version of DisplacementControl integrator. In an analysis step with Displacement Control we
seek to determine the time step that will result in a displacement increment for a particular degree-of-freedom
at a node to be a prescribed value.

nodeTag (int) tag of node whose response controls solution

dof (int) Degree of freedom at the node, 1 through ndf.
incr (float) First displacement increment ∆𝑈𝑑𝑜𝑓 .
numIter (int) Number of iterations the user would like to occur in the solution algorithm. (optional)
minIncr (float) Min stepsize the user will allow ∆𝑈𝑚𝑖𝑛 . (optional)
maxIncr (float) Max stepsize the user will allow ∆𝑈𝑚𝑎𝑥 . (optional)

Use this command only for parallel model.

Warning: Don’t use this command if model is not parallel, for example, parametric study.

Minimum Unbalanced Displacement Norm

integrator(’MinUnbalDispNorm’, dlambda1, Jd=1, minLambda=dlambda1, maxLambda=dlambda1,

det=False)
Create a MinUnbalDispNorm integrator.

dlambda1 First load increment (pseudo-time step) at the first iteration in the next invocation of the
(float) analysis command.
Jd (int) Factor relating first load increment at subsequent time steps. (optional)
minLambda Min load increment. (optional)
(float)
maxLambda Max load increment. (optional)
(float)

Arc-Length Control

integrator(’ArcLength’, s, alpha)
Create a ArcLength integrator. In an analysis step with ArcLength we seek to determine the time step that will
result in our constraint equation being satisfied.

s (float) The arcLength.

alpha (float) 𝛼 a scaling factor on the reference loads.

Transient integrator objects

1. Central Difference
2. Newmark Method
3. Hilber-Hughes-Taylor Method

240 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

4. Generalized Alpha Method

5. TRBDF2
6. Explicit Difference
7. PFEM integrator

Central Difference

integrator(’CentralDifference’)
Create a centralDifference integrator.
1. The calculation of 𝑈𝑡 +∆𝑡, is based on using the equilibrium equation at time t. For this reason the method
is called an explicit integration method.
2. If there is no rayleigh damping and the C matrix is 0, for a diagonal mass matrix a diagonal solver may
and should be used.
Δ𝑡 1
3. For stability, 𝑇𝑛 < 𝜋

Newmark Method

integrator(’Newmark’, gamma, beta, ’-form’, form)

Create a Newmark integrator.

gamma (float) 𝛾 factor.

beta (float) 𝛽 factor.
form (str) Flag to indicate which variable to be used as primary
variable (optional)
• 'D' – displacement (default)
• 'V' – velocity
• 'A' – acceleration

1. If the accelerations are chosen as the unknowns and 𝛽 is chosen as 0, the formulation results in the fast but
conditionally stable explicit Central Difference method. Otherwise the method is implicit and requires an
iterative solution process.
2. Two common sets of choices are
1. Average Acceleration Method (𝛾 = 21 , 𝛽 = 14 )
2. Linear Acceleration Method (𝛾 = 12 , 𝛽 = 16 )
1 1
3. 𝛾 > 2 results in numerical damping proportional to 𝛾 − 2
1
4. The method is second order accurate if and only if 𝛾 = 2
𝛾 1
5. The method is unconditionally stable for 𝛽 >= 2 >= 4

Hilber-Hughes-Taylor Method

integrator(’HHT’, alpha, gamma=1.5-alpha, beta=(2-alpha)^2/4)

Create a Hilber-Hughes-Taylor (HHT) integrator. This is an implicit method that allows for energy dissipation

1.5. Analysis Commands 241

OpenSeesPy Documentation, Release 3.5.1.3

and second order accuracy (which is not possible with the regular Newmark object). Depending on choices of
input parameters, the method can be unconditionally stable.

alpha (float) 𝛼 factor.

gamma (float) 𝛾 factor. (optional)
beta (float) 𝛽 factor. (optional)

1. Like Mewmark and all the implicit schemes, the unconditional stability of this method applies to linear
problems. There are no results showing stability of this method over the wide range of nonlinear problems
that potentially exist. Experience indicates that the time step for implicit schemes in nonlinear situations
can be much greater than those for explicit schemes.
2. 𝛼 = 1.0 corresponds to the Newmark method.
3. 𝛼 should be between 0.67 and 1.0. The smaller the 𝛼 the greater the numerical damping.
4. 𝛾 and 𝛽 are optional. The default values ensure the method is second order accurate and unconditionally
stable when 𝛼 is 23 <= 𝛼 <= 1.0. The defaults are:
(2−𝛼)2
𝛽= 4

and
3
𝛾= 2 −𝛼

Generalized Alpha Method

integrator(’GeneralizedAlpha’, alphaM, alphaF, gamma=0.5+alphaM-alphaF, beta=(1+alphaM-

alphaF)^2/4)
Create a GeneralizedAlpha integrator. This is an implicit method that like the HHT method allows for high
frequency energy dissipation and second order accuracy, i.e. ∆𝑡2 . Depending on choices of input parameters,
the method can be unconditionally stable.

alphaM (float) 𝛼𝑀 factor.

alphaF (float) 𝛼𝐹 factor.
gamma (float) 𝛾 factor. (optional)
beta (float) 𝛽 factor. (optional)

1. Like Newmark and all the implicit schemes, the unconditional stability of this method applies to linear
problems. There are no results showing stability of this method over the wide range of nonlinear problems
that potentially exist. Experience indicates that the time step for implicit schemes in nonlinear situations
can be much greater than those for explicit schemes.
2. 𝛼𝑀 = 1.0, 𝛼𝐹 = 1.0 produces the Newmark Method.
3. 𝛼𝑀 = 1.0 corresponds to the integrator.HHT() method.
1
4. The method is second-order accurate provided 𝛾 = 2 + 𝛼𝑀 − 𝛼𝐹
5. The method is unconditionally stable provided 𝛼𝑀 >= 𝛼𝐹 >= 21 , 𝛽 >= 1
4 + 12 (𝛾𝑀 − 𝛾𝐹 )
6. 𝛾 and 𝛽 are optional. The default values ensure the method is unconditionally stable, second order accurate
and high frequency dissipation is maximized.
The defaults are:
1
𝛾= 2 + 𝛼𝑀 − 𝛼𝐹
and

242 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

𝛽 = 14 (1 + 𝛼𝑀 − 𝛼𝐹 )2

TRBDF2

integrator(’TRBDF2’)
Create a TRBDF2 integrator. The TRBDF2 integrator is a composite scheme that alternates between
the Trapezoidal scheme and a 3 point backward Euler scheme. It does this in an attempt to conserve
energy and momentum, something Newmark does not always do.
As opposed to dividing the time-step in 2 as outlined in the Bathe2007, we just switch alternate between the 2
integration strategies,i.e. the time step in our implementation is double that described in the Bathe2007.

Explicit Difference

integrator(’ExplicitDifference’)
Create a ExplicitDifference integrator.
1. When using Rayleigh damping, the damping ratio of high vibration modes is overrated, and the critical
time step size will be much smaller. Hence Modal damping is more suitable for this method.
2. There should be no zero element on the diagonal of the mass matrix when using this method.
3. Diagonal solver should be used when lumped mass matrix is used because the equations are uncoupled.
(︁√︀ )︁
4. For stability, ∆𝑡 ≤ 𝜁 2 + 1 − 𝜁 𝜔2

1.5.7 analysis command

analysis(analysisType)
This command is used to construct the Analysis object, which defines what type of analysis is to be performed.
• determine the predictive step for time t+dt
• specify the tangent matrix and residual vector at any iteration
• determine the corrective step based on the displacement increment dU

analysisType (str) char string identifying type of analysis object to be

constructed. Currently 3 valid options:
1. 'Static' - for static analysis
2. 'Transient' - for transient analysis con-
stant time step
3. 'VariableTransient' - for transient
analysis with variable time step
4. 'PFEM' - for PFEM analysis.

Note: If the component objects are not defined before hand, the command automatically creates default component
objects and issues warning messages to this effect. The number of warning messages depends on the number of
component objects that are undefined.

1.5. Analysis Commands 243

OpenSeesPy Documentation, Release 3.5.1.3

1.5.8 eigen command

eigen(solver=’-genBandArpack’, numEigenvalues)
Eigen value analysis. Return a list of eigen values.

numEigenvalues number of eigenvalues required

(int)
solver (str) optional string detailing type of solver: '-genBandArpack',
'-fullGenLapack', (optional)

Note:
1. The eigenvectors are stored at the nodes and can be printed out using a Node Recorder, the nodeEigenvector
command, or the Print command.
2. The default eigensolver is able to solve only for N-1 eigenvalues, where N is the number of inertial DOFs. When
running into this limitation the -fullGenLapack solver can be used instead of the default Arpack solver.
3. The -fullGenLapack option is VERY SLOW for moderate to large models

1.5.9 analyze command

analyze(numIncr=1, dt=0.0, dtMin=0.0, dtMax=0.0, Jd=0)

Perform the analysis. Return 0 if successful, <0 if NOT successful

numIncrNumber of analysis steps to perform. (required except for PFEM analysis)

(int)
dt Time-step increment. (required for Transient analysis and VariableTransient analysis.‘)
(float)
dtMin Minimum time steps. (required for VariableTransient analysis)
(float)
dtMax Maximum time steps (required for VariableTransient analysis)
(float)
Jd Number of iterations user would like performed at each step. The variable transient analysis will
(float) change current time step if last analysis step took more or less iterations than this to converge
(required for VariableTransient analysis)

1.5.10 modalProperties Command

This command is used to compute the modal properties of a model after an eigen command.
This command computes the following modal properties:

• 𝑚𝑡 : A vector with the total mass of the structure for each DOF, considering both free and fixed DOFs.
• 𝑚𝑓 : The total mass of the structure, but considering only the masses at free DOFs.
• 𝐶𝑜𝑀 : The center of mass.
• 𝑔𝑚 : The generalized mass matrix.
• 𝑀 𝑃 𝐹 : The modal participation factors.

244 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

• 𝑀 𝑃 𝑀 : The modal participation masses.

• 𝑀 𝑃 𝑀 𝑐 : The cumulative modal participation masses.
• 𝑀 𝑃 𝑀 (%) : The modal participation masse ratios.
• 𝑀 𝑃 𝑀 𝑐 (%) : The cumulative modal participation masse ratios.
modalProperties(<’-print’>, <’-file’, reportFileName>, <’-unorm’>, <’-return’>)

Argument Type Description

'-print' (str) Optional. If included, a report of the modal properties is printed to the
console.
'-file' (str) Optional. If included, a report of the modal properties is printed to the file
reportFileName.
(str)
reportFileName Optional, but mandatory if the -file option is included. Indicates the file-
name for the report. If the file does not exist, it will be created. If the file
exists, it will be overwritten.
'-unorm' (str) Optional. If included, the computation of the modal properties will be car-
ried out using a displacement-normalized version of the eigenvectors.
'-return' (str) Optional. If included, a report of the modal properties will be returned as a
dict object to Python.

Return value with '-return' (In next version 3.4.0.3 or later)

{
# list of one int value
"domainSize": [ndm],

# list of lambda values for all modes

"eigenLambda": [lambdas],

# list of omega values for all modes

"eigenOmega": [omega],

# list of frequency values for all modes

"eigenFrequency": [frequency],

# list of period values for all modes

"eigenPeriod": [period],

# list of total mass values, [MX] for 1D, [MX, MY, RMZ] for 2D,
# [MX, MY, MZ, MRX, RMY, RMZ] for 3D
"totalMass": [mass],

# list of total mass values for free DOFs, [MX] for 1D,
# [MX, MY, RMZ] for 2D, [MX, MY, MZ, MRX, RMY, RMZ] for 3D
"totalFreeMass": [mass],

# coordinates of mass center, [X] for 1D, [X, Y] for 2D,

# [X, Y, Z] for 3D
"centerOfMass": [center],

#
# modal participation factors for all modes
#
(continues on next page)

1.5. Analysis Commands 245

OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

# for MX direction for 1D, 2D and 3D

"partiFactorMX": [factor],

# for MY direction for 2D and 3D

"partiFactorMY": [factor],

# for MZ direction for 3D

"partiFactorMZ": [factor],

# for RMX direction for 3D

"partiFactorRMX": [factor],

# for RMY direction for 3D

"partiFactorRMY": [factor],

# for RMZ direction for 2D and 3D

"partiFactorRMX": [factor],

#
# modal participation masses for all modes
#

# for MX direction for 1D, 2D and 3D

"partiMassMX": [mass],

# for MY direction for 2D and 3D

"partiMassMY": [mass],

# for MZ direction for 3D

"partiMassMZ": [mass],

# for RMX direction for 3D

"partiMassRMX": [mass],

# for RMY direction for 3D

"partiMassRMY": [mass],

# for RMZ direction for 2D and 3D

"partiMassRMX": [mass],

#
# modal participation masses (cumulative) for all modes
#

# for MX direction for 1D, 2D and 3D

"partiMassesCumuMX": [mass],

# for MY direction for 2D and 3D

"partiMassesCumuMY": [mass],

# for MZ direction for 3D

"partiMassesCumuMZ": [mass],

# for RMX direction for 3D

"partiMassesCumuRMX": [mass],

(continues on next page)

246 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

# for RMY direction for 3D
"partiMassesCumuRMY": [mass],

# for RMZ direction for 2D and 3D

"partiMassesCumuRMZ": [mass],

#
# modal participation mass ratios (%) for all modes
#

# for MX direction for 1D, 2D and 3D

"partiMassRatiosMX": [mass],

# for MY direction for 2D and 3D

"partiMassRatiosMY": [mass],

# for MZ direction for 3D

"partiMassRatiosMZ": [mass],

# for RMX direction for 3D

"partiMassRatiosRMX": [mass],

# for RMY direction for 3D

"partiMassRatiosRMY": [mass],

# for RMZ direction for 2D and 3D

"partiMassRatiosRMX": [mass],

#
# modal participation mass ratios (%) (cumulative) for all modes
#

# for MX direction for 1D, 2D and 3D

"partiMassRatiosCumuMX": [mass],

# for MY direction for 2D and 3D

"partiMassRatiosCumuMY": [mass],

# for MZ direction for 3D

"partiMassRatiosCumuMZ": [mass],

# for RMX direction for 3D

"partiMassRatiosCumuRMX": [mass],

# for RMY direction for 3D

"partiMassRatiosCumuRMY": [mass],

# for RMZ direction for 2D and 3D

"partiMassRatiosCumuRMX": [mass],
}

Note:
• This command can be used only if a previous call to eigen command has been performed.
• This command has only optional arguments, and they can be given in any order. Note, however, that the only

1.5. Analysis Commands 247

OpenSeesPy Documentation, Release 3.5.1.3

requirement is that the reportFileName must follow the '-file' option if used.
• The modal properties are computed and stored in the Domain object, so that they can be accessed later in the
responseSpectrumAnalysis Command.
• This command directly accesses the mass matrix of the model, so it accounts for both nodal masses and element’s
(distributed) masses. And the element mass matrix can be either lumped or consistent.
• The global mass matrix is then stored into a temporary sparse matrix storage, so it can be used for both small
and large models.
• This command can be used for both 2D problems (-ndm = 2, -ndf = 2 or 3) and 3D problems (-ndm = 3, -ndf =
3, 4 or 6). In both cases the algorithm computes rotational masses. If a node has rotational DOFs, the rotational
masses account for both the direct rotational masses (i.e. those input by the user at rotational DOFs) and the
effect of translational masses gyrating about the center of mass. If a node has no rotational DOF, only the latter
is considered.
• Only the values in 𝑔𝑚 and 𝑀 𝑃 𝐹 depend on the normalization of the eigenvectors. This normalization depends
on the solver used in the eigen command. The default '-genBandArpack' uses a mass-normalization, so that
the 𝑔𝑚 is the identity. On the contrary, the '-fullGenLapack' uses a displacement-normalization, so that
the largest component of the eigenvector is 1. If you use the '-genBandArpack', but want a displacement-
normalization of the eigenvectors, use the '-unorm' option.

Theory

The eigenvalues 𝜆 and the eigenvectors Φ can be obtained after solving the generalized eigenvalue problem for two
symmetric matrices 𝐾 (stiffness) and 𝑀 (mass) given by:

(𝐾 − 𝜆𝑀 ) Φ = 0

The global mass matrix 𝑀 is given by the assembly of 𝑛 elemental and nodal mass matrices 𝑚𝑒 :
𝑛
⋀︁
𝑀= 𝑚𝑒
𝑖=1

𝑀 is not necessarily diagonal, because some elemental matrices 𝑚𝑒 may be consistent. However, the computation of
𝐶𝑜𝑀 , 𝑚𝑡 and 𝑚𝑓 requires a lumped version of 𝑀 . The global lumped mass matrix 𝐿𝑀 can be computed by the
assembly of a diagonalized version of the elemental mass matrices 𝑚𝑒 :
𝑛
⋀︁
𝐿𝑀 = 𝑑𝑖𝑎𝑔 (𝑚𝑒 )
𝑖=1

𝑑𝑖𝑎𝑔 (𝑚𝑒 ) cannot be computed just by summing the summing the components of each row (in beams or solid with
higher order interpolation, this would produces negative terms on the diagonal mass matrix that would be unphysical).
Instead we use the HRZ algorithm [HintonEtAl1976], named after the authors Hinton, Rock and Zienkiewicz: “The
procedure of lumping recommended in view of the infinite possibilities offered by condition (5) is to compute the
diagonal terms of the consistent mass matrix and then scale these terms so as to preserve the total mass of the
element”.
The procedure is as follows:
• compute 𝐷𝑀 , a vector containing the sum of each row of 𝑚𝑒 .

248 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

• compute 𝑆𝑀 , a vector of size=ndf, obtained summing the components in 𝐷𝑀 pertaining to the same DOF (i
to ndf). This procedures allows to obtain the total elemental mass for each DOF.
• compute 𝐷𝐶, a vector containing only the diagonal terms in the consistent mass matrix 𝑚𝑒 .
• compute 𝑆𝐶, a vector of size=ndf, obtained summing the components in 𝐷𝐶 pertaining to the same DOF (i to
ndf).
• now we can obtain the scale factors for each dof 𝑖 as: 𝑆𝑀𝑖 /𝑆𝐶𝑖 .
• scale each diagonal term of the consistent mass matrix 𝐷𝐶𝑗 using the scale factor of the respective DOF
𝑆𝑀𝑖 /𝑆𝐶𝑖 :
𝑑𝑖𝑎𝑔 (𝑚𝑒 )𝑗 = 𝐷𝐶𝑗 · 𝑆𝑀𝑖 /𝑆𝐶𝑖 .

The center of mass 𝐶𝑜𝑀 and the total masses 𝑚𝑡 and 𝑚𝑓 of the structure, for each node 𝑛 with position 𝑋𝑛 and
each DOF 𝑖, can now be easily computed from 𝐿𝑀 :

𝑁 ∑︁
𝑛𝑜𝑑𝑒𝑠
𝑚𝑡𝑖 = 𝐿𝑀𝑛𝑖
𝑛=1
𝑁 ∑︁
𝑛𝑜𝑑𝑒𝑠
𝑚𝑓𝑖 = 𝐿𝑀𝑛𝑖 (if 𝑖 = 𝑓 𝑟𝑒𝑒)
𝑛=1
∑︀𝑁 𝑛𝑜𝑑𝑒𝑠
𝑛=1 𝑋𝑛𝑖 · 𝐿𝑀𝑛𝑖
𝐶𝑜𝑀𝑖 = (if 𝑖 = 𝑓 𝑟𝑒𝑒)
𝑚𝑓𝑖

The generalized mass matrix is

𝑔𝑚 = Φ𝑇 𝑀 Φ

If the default solver is used in the eigen command (-genBandArpack), and the option -unorm is not used, the
eigenvectors are mass-normalized and 𝑔𝑚 will be an identity matrix, i.e. a diagonal matrix whose diagonal entries
are = 1, and whose size is 𝑛𝑚 × 𝑛𝑚 (where 𝑛𝑚 is the number of requested eigenvalues).

The modal participation factor matrix 𝑀 𝑃 𝐹 is a 𝑛𝑚 × 𝑛𝑑𝑓 matrix (where ndf = 3 in 2D and 6 in 3D), where each
row contains the modal participation factors for each DOF. The modal participation factor for a certain mode 𝑖 and
DOF 𝑗 indicates how strongly the motion (or rotation) associated to that DOF is represented in the eigenvector 𝑖

Φ𝑇𝑖 𝑀 𝑇𝑗
𝑀 𝑃 𝐹𝑖𝑗 =
𝑔𝑚𝑖𝑖

where 𝑇𝑗 defines the magnitude of the rigid body response to imposed rigid body motion (displacement or
infinitesimal rotation) in the DOF 𝑗. Each 𝑛𝑑𝑓 × 1 block 𝑇𝑛𝑗 corresponds to the node 𝑛 and it is defined as (for the
3D/6DOFs case):
⎛ ⎞⎧ ⎫
1 0 0 0 𝑑𝑧 −𝑑𝑧 ⎪ ⎪ 𝑒1 ⎪
⎪
⎜0 1 0 −𝑑𝑧 0 𝑑 𝑥 ⎟⎪
⎟ ⎪
⎪ 𝑒2 ⎪
⎪
⎪
⎜ ⎪ ⎪
⎜0 0 1 𝑑𝑦 −𝑑𝑥 0 ⎟
⎨
𝑒3
⎬
𝑇𝑛𝑗 = ⎜⎜ ⎟
⎜0 0 0 1 0 0 ⎟⎟⎪⎪ 𝑒4 ⎪
⎪ ⎪
⎝0 0 0 0 1 0 ⎠⎪ ⎪ 𝑒5 ⎪
⎪
⎪
⎪
⎩ ⎪
0 0 0 0 0 1 𝑒6
⎭

1.5. Analysis Commands 249

OpenSeesPy Documentation, Release 3.5.1.3

where 𝑒𝑗 is 1, and all other 𝑒𝑝 (with 𝑝 ̸= 𝑗) are 0. 𝑑𝑥 , 𝑑𝑦 and 𝑑𝑧 are the distances of the node 𝑛 coordinates
𝑋𝑛 = (𝑥, 𝑦, 𝑧) from the center of mass 𝐶𝑜𝑀 = (𝑥0 , 𝑦0 , 𝑧0 ). Therefore, the modal participation factors accounts for
the masses directly input at translational and rotational DOFs, and also the rotational masses given by the
translational masses gyrating about the center of mass. Note, in fact, that even if the user does not input any rotational
mass, or even if the user uses 3D solid elements with no rotational DOF, the modal participation factors associated to
the rotational DOFs may be ̸= 0.

The modal participation mass matrix 𝑀 𝑃 𝑀 is a 𝑛𝑚 × 𝑛𝑑𝑓 matrix (where ndf = 3 in 2D and 6 in 3D), where each
row contains the modal participation masses for each DOF. The modal participation mass for a certain mode 𝑖 and
DOF 𝑗 is defined as
(︀ 𝑇 )︀2
Φ 𝑖 𝑀 𝑇𝑗
𝑀 𝑃 𝑀𝑖𝑗 =
𝑔𝑚𝑖𝑖

If the modal participation masses for each mode in a particular DOF are summed, it should give the total mass of the
structure for that DOF, exlcluding the masses at fixed DOFs.

Example
The following example shows how to:
• Use the modalProperties command
• Print the results on the console ('-print')
• Generate a report file in the current directory ('-file', 'ModalReport.txt')
• Use a displacement-normalization for the eigenvectors

modalProperties('-print', '-file', 'ModalReport.txt', '-unorm')

For a complete example that runs an eigenvalue analysis, extracts the modal properties and runs a response spec-
trum analysis, see the documentation of the responseSpectrumAnalysis Command

ReportFile
The generated report file looks like this:

# MODAL ANALYSIS REPORT

* 1. DOMAIN SIZE:
# This is the size of the problem: 2 for 2D problems, 3 for 3D problems.
3

* 2. EIGENVALUE ANALYSIS:
# MODE LAMBDA OMEGA FREQUENCY PERIOD
# ------------- ------------- ------------- ------------- -------------
1 7578.8 87.0563 13.8554 0.0721738
2 8484.47 92.1112 14.6599 0.0682131
3 10518.5 102.56 16.3229 0.0612636
4 85779 292.881 46.6134 0.0214531
5 89260.1 298.764 47.5498 0.0210306
6 101089 317.945 50.6025 0.0197619
(continues on next page)

250 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

7 1.71885e+06 1311.05 208.66 0.00479249

* 3. TOTAL MASS OF THE STRUCTURE:

# The total masses (translational and rotational) of the structure
# including the masses at fixed DOFs (if any).
# MX MY MZ RMX RMY RMZ
# ------------- ------------- ------------- ------------- ------------- -------------
1600 1600 1600 7200 10000 10000

* 4. TOTAL FREE MASS OF THE STRUCTURE:

# The total masses (translational and rotational) of the structure
# including only the masses at free DOFs.
# MX MY MZ RMX RMY RMZ
# ------------- ------------- ------------- ------------- ------------- -------------
1600 1600 1600 7200 10000 10000

* 5. CENTER OF MASS:
# The center of mass of the structure, calculated from free masses.
# X Y Z
# ------------- ------------- -------------
2 1.5 4.5

* 6. MODAL PARTICIPATION FACTORS:

# The participation factor for a certain mode 'a' in a certain direction 'i'
# indicates how strongly displacement along (or rotation about)
# the global axes is represented in the eigenvector of that mode.
# MODE MX MY MZ RMX RMY
˓→ RMZ
# ------------- ------------- ------------- ------------- ------------- -------------
˓→-------------

1 1.20368 0 0 0 0.661418
˓→ 0
2 0 -1.20172 0 0.637456 0
˓→ 0
3 0 0 0 0 0
˓→ -2.39705
4 0.430981 0 0 0 -1.8352
˓→ 0
5 0 -0.41375 0 -1.83591 0
˓→ 0
6 0 0 0 0 0
˓→ 0.780575
7 0 0 -1.17082 0 0
˓→ 0

* 7. MODAL PARTICIPATION MASSES:

# The modal participation masses for each mode.
# MODE MX MY MZ RMX RMY
˓→ RMZ
# ------------- ------------- ------------- ------------- ------------- -------------
˓→-------------

1 1418.18 0 0 0 428.21
˓→ 0 (continues on next page)

1.5. Analysis Commands 251

OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

2 0 1430.41 0 402.49 0
˓→ 0
3 0 0 0 0 0
˓→ 9041.23
4 181.82 0 0 0 3296.78
˓→ 0
5 0 169.58 0 3338.87 0
˓→ 0
6 0 0 0 0 0
˓→ 958.755
7 0 0 1515.54 0 0
˓→ 0

* 8. MODAL PARTICIPATION MASSES (cumulative):

# The cumulative modal participation masses for each mode.
# MODE MX MY MZ RMX RMY
˓→ RMZ
# ------------- ------------- ------------- ------------- ------------- -------------
˓→-------------

1 1418.18 0 0 0 428.21
˓→ 0
2 1418.18 1430.41 0 402.49 428.21
˓→ 0
3 1418.18 1430.41 0 402.49 428.21
˓→ 9041.23
4 1600 1430.41 0 402.49 3724.99
˓→ 9041.23
5 1600 1599.99 0 3741.36 3724.99
˓→ 9041.23
6 1600 1599.99 0 3741.36 3724.99
˓→ 9999.99
7 1600 1599.99 1515.54 3741.36 3724.99
˓→ 9999.99

* 9. MODAL PARTICIPATION MASS RATIOS (%):

# The modal participation mass ratios (%) for each mode.
# MODE MX MY MZ RMX RMY
˓→ RMZ
# ------------- ------------- ------------- ------------- ------------- -------------
˓→-------------

1 88.636 0 0 0 4.2821
˓→ 0
2 0 89.4005 0 5.59014 0
˓→ 0
3 0 0 0 0 0
˓→ 90.4123
4 11.3638 0 0 0 32.9678
˓→ 0
5 0 10.5988 0 46.3732 0
˓→ 0
6 0 0 0 0 0
˓→ 9.58755
7 0 0 94.7214 0 0
˓→ 0

(continues on next page)

252 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

* 10. MODAL PARTICIPATION MASS RATIOS (%) (cumulative):

# The cumulative modal participation mass ratios (%) for each mode.
# MODE MX MY MZ RMX RMY
˓→ RMZ
# ------------- ------------- ------------- ------------- ------------- -------------
˓→-------------

1 88.636 0 0 0 4.2821
˓→ 0
2 88.636 89.4005 0 5.59014 4.2821
˓→ 0
3 88.636 89.4005 0 5.59014 4.2821
˓→ 90.4123
4 99.9997 89.4005 0 5.59014 37.2499
˓→ 90.4123
5 99.9997 99.9993 0 51.9633 37.2499
˓→ 90.4123
6 99.9997 99.9993 0 51.9633 37.2499
˓→ 99.9999
7 99.9997 99.9993 94.7214 51.9633 37.2499
˓→ 99.9999

Code Developed by: Massimo Petracca at ASDEA Software, Italy

1.5.11 responseSpectrumAnalysis Command

This command is used to perform a response spectrum analysis.

The response spectrum analysis performs N linear analysis steps, where N is the number of eigenvalues requested in
a previous call to eigen command.
For each analysis step, it computes the modal displacements. When the i-th analysis step is complete, all previously
defined recorders will be called, so they will record all the results requested by the user, pertaining to the current
modal displacements.
The modal combination of these modal displacements (and derived results such as beam forces) is up to the user, and
can be easily done via Python scripting.

The command can be called in two different ways, depending on how you store the Tn/Sa (response spectrum
function) values.
They can be either stored in a timeSeries . . .

responseSpectrumAnalysis(tsTag, direction, <’-scale’, scale>, <’-mode’, mode>)

. . . or in two lists

responseSpectrumAnalysis(direction, ’-Tn’, Tn, ’-Sa’, Sa, <’-scale ’, scale>, <’-mode’, mode>)

1.5. Analysis Commands 253

OpenSeesPy Documentation, Release 3.5.1.3

Argument Type Description

tsTag (int) The tag of a previously defined timeSeries commands. This list stores the
Tn and Sa values. If you want to use the timeSeries, you cannot specify the
'-Tn' and '-Sa' options
'-Tn' (str) Tells the command to use the Tn list, instead of the timeSeries, to get the
periods of the response spectrum function
Tn (list (float)) The list of periods of the response spectrum function
'-Sa' (str) Tells the command to use the Sa list, instead of the timeSeries, to get the
accelerations of the response spectrum function
Sa (list (float)) The list of accelerations of the response spectrum function
direction (int) The 1-based index of the excited DOF (1 to 3 for 2D problems, or 1 to 6 for
3D problems).
'-scale' (str) Tells the command to use a user-defined scale factor for the computed modal
displacements. Not used, placeholder for future implementation.
scale (float) User-defined scale factor for the computed modal displacements. Not used,
placeholder for future implementation.
'-mode' (str) Tells the command to compute the modal displacements for just 1 specified
mode (by default all modes are processed).
mode (int) The 1-based index of the unique mode to process.

Note:
• This command can be used only if a previous call to eigen command and modalProperties Command has been
performed.
• It computes only the modal displacements, any modal combination is up to the user.
• The scale factor ('-scale', scale) for the output modal displacements is not used now, and it’s there for
future implementations. When your model is linear elastic there is no need to use this option. It will be useful in
future when we will allow using this command on a nonlinear model as a linear perturbation about a certain
nonlinear state. In that case, the scale factor can be set to a very small number, so that the computed modal
displacements will be very small (linear perturbation) and will not alter the nonlinear state of your model. Then
the inverse of the scale factor can be used to post-multiply any result for post-processing.

Theory

Once the eigenvalue problem (eigen command) has been solved, and once the modal properties (modalProperties
Command) have been computed, the modal displacements 𝑈 for mode 𝑖 at node 𝑛 and DOF 𝑗 is given by

𝑛
Φ𝑛𝑖𝑗 · 𝑀 𝑃 𝐹𝑖𝑗 · 𝑅𝑆𝑓 (𝑇𝑖 )
𝑈𝑖𝑗 =
𝜆𝑖

where:
• 𝜆𝑖 is the eigenvalue
• Φ𝑛𝑖𝑗 is the eigenvector
• 𝑀 𝑃 𝐹𝑖𝑗 is the modal participation factor
• 𝑅𝑆𝑓 (𝑇𝑖 ) is the response spectrum function value at period 𝑇𝑖

254 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

2𝜋
• and 𝑇𝑖 is the period √
𝜆𝑖

Example 1: Simple call

The following example shows how to call the responseSpectrumAnalysis command for all modes, using the time series
1 (or lists Tn and Sa) along the DOF 1 (Ux)
1. Using timeSeries

tsTag = 1 # use the timeSeries 1 as response spectrum function

direction = 1 # excited DOF = Ux
responseSpectrumAnalysis(tsTag, direction)

2. Using lists

Tn = [0.0 0.1 0.4 .... ] # the periods

Sa = [1.9 3.7 4.9 .... ] # the accelerations
responseSpectrumAnalysis(direction, '-Tn', *Tn, '-Sa', *Sa)

Example 2: Iterative call

The following example shows how to call the responseSpectrumAnalysis command for 1 mode at a time, using the
time series 1 along the DOF 1 (Ux)

tsTag = 1 # use the timeSeries 1 as response spectrum function

direction = 1 # excited DOF = Ux
for i in range(num_modes):
responseSpectrumAnalysis(tsTag, direction, '-mode', i+1)
# grab your results here for the i-th modal displacements

Example 3: Complete Structural Example

The following example show a simple 1-bay 2-story building with rigid diaphragms. Units are Newton and meters.
It shows how to:

• call the eigen command to extract 7 modes of vibration

• call the modalProperties Command to generate the report with modal properties

1.5. Analysis Commands 255

OpenSeesPy Documentation, Release 3.5.1.3

• call the responseSpectrumAnalysis Command to compute the modal displacements and section forces * in a first
example the responseSpectrumAnalysis Command is called for all modes. Results are obtained from a recorder
after the analysis. * in a second example the responseSpectrumAnalysis Command is called in a for-loop mode-
by-mode. Results are obtained within the for-loop usin the eleResponse command
• do a CQC modal combination

responseSpectrumAnalysisExample.py

Code Developed by: Massimo Petracca at ASDEA Software, Italy

1.6 Output Commands

Get outputs from OpenSees. These commands don’t change internal states of OpenSees.
1. basicDeformation command
2. basicForce command
3. basicStiffness command
4. eleDynamicalForce command
5. eleForce command
6. eleNodes command
7. eleResponse command
8. getEleTags command
9. getLoadFactor command
10. getNodeTags command
11. getTime command
12. nodeAccel command
13. nodeBounds command
14. nodeCoord command
15. nodeDisp command
16. nodeEigenvector command
17. nodeDOFs command
18. nodeMass command
19. nodePressure command
20. nodeReaction command
21. nodeResponse command
22. nodeVel command
23. nodeUnbalance command
24. numFact command

256 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

25. numIter command

26. printA command
27. printB command
28. printGID command
29. printModel command
30. record command
31. recorder command
32. sectionForce command
33. sectionDeformation command
34. sectionStiffness command
35. sectionFlexibility command
36. sectionLocation command
37. sectionWeight command
38. sectionTag
39. systemSize command
40. testIter command
41. testNorm command
42. version command
43. logFile command

1.6.1 basicDeformation command

basicDeformation(eleTag)
Returns the deformation of the basic system for a beam-column element.

eleTag (int) element tag.

1.6.2 basicForce command

basicForce(eleTag)
Returns the forces of the basic system for a beam-column element.

eleTag (int) element tag.

1.6.3 basicStiffness command

basicStiffness(eleTag)
Returns the stiffness of the basic system for a beam-column element. A list of values in row order will be
returned.

eleTag (int) element tag.

1.6. Output Commands 257

OpenSeesPy Documentation, Release 3.5.1.3

1.6.4 eleDynamicalForce command

eleDynamicalForce(eleTag, dof=-1)
Returns the elemental dynamic force.

eleTag element tag.

(int)
dof (int) specific dof at the element, (optional), if no dof is provided, a list of values for all dofs is
returned.

1.6.5 eleForce command

eleForce(eleTag, dof=-1)
Returns the elemental resisting force.

eleTag element tag.

(int)
dof (int) specific dof at the element, (optional), if no dof is provided, a list of values for all dofs is
returned.

1.6.6 eleNodes command

eleNodes(eleTag)
Get nodes in an element

eletag (int) element tag.

1.6.7 eleResponse command

eleResponse(eleTag, *args)
This command is used to obtain the same element quantities as those obtained from the element recorder at a
particular time step.

eletag element tag.

(int)
args same arguments as those specified in element recorder. These arguments are specific to the
(list) type of element being used.

1.6.8 getEleTags command

getEleTags(’-mesh’, mtag)
Get all elements in the domain or in a mesh.

mtag (int) mesh tag. (optional)

258 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1.6.9 getLoadFactor command

getLoadFactor(patternTag)
Returns the load factor 𝜆 for the pattern

patternTag (int) pattern tag.

1.6.10 getNodeTags command

getNodeTags(’-mesh’, mtag)
Get all nodes in the domain or in a mesh.

mtag (int) mesh tag. (optional)

1.6.11 getTime command

getTime()
Returns the current time in the domain.

1.6.12 nodeAccel command

nodeAccel(nodeTag, dof=-1)
Returns the current acceleration at a specified node.

nodeTag node tag.

(int)
dof (int) specific dof at the node (1 through ndf), (optional), if no dof is provided, a list of values for
all dofs is returned.

1.6.13 nodeBounds command

nodeBounds()
Get the boundary of all nodes. Return a list of boundary values.

1.6.14 nodeCoord command

nodeCoord(nodeTag, dim=-1)
Returns the coordinates of a specified node.

nodeTag node tag.

(int)
dof (int) specific dimension at the node (1 through ndf), (optional), if no dim is provided, a list of
values for all dimensions is returned.

1.6. Output Commands 259

OpenSeesPy Documentation, Release 3.5.1.3

1.6.15 nodeDisp command

nodeDisp(nodeTag, dof=-1)
Returns the current displacement at a specified node.

nodeTag node tag.

(int)
dof (int) specific dof at the node (1 through ndf), (optional), if no dof is provided, a list of values for
all dofs is returned.

1.6.16 nodeEigenvector command

nodeEigenvector(nodeTag, eigenvector, dof=-1)

Returns the eigenvector at a specified node.

nodeTag (int) node tag.

eigenvector mode number of eigenvector to be returned
(int)
dof (int) specific dof at the node (1 through ndf), (optional), if no dof is provided, a list of values
for all dofs is returned.

1.6.17 nodeDOFs command

nodeDOFs(nodeTag)
Returns the DOF numbering of a node.

nodeTag (int) node tag.

1.6.18 nodeMass command

nodeMass(nodeTag, dof=-1)
Returns the mass at a specified node.

nodeTag node tag.

(int)
dof (int) specific dof at the node (1 through ndf), (optional), if no dof is provided, a list of values for
all dofs is returned.

1.6.19 nodePressure command

nodePressure(nodeTag)
Returns the fluid pressures at a specified node if this is a fluid node.

nodeTag (int) node tag.

260 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1.6.20 nodeReaction command

nodeReaction(nodeTag, dof=-1)
Returns the reactions at a specified node. Must call reactions() command before this command.

nodeTag node tag.

(int)
dof (int) specific dof at the node (1 through ndf), (optional), if no dof is provided, a list of values for
all dofs is returned.

1.6.21 nodeResponse command

nodeResponse(nodeTag, dof, responseID)

Returns the responses at a specified node. To get reactions (id=6), must call the reactions command before
this command.

nodeTag (int) node tag.

dof (int) specific dof of the response
responseID (int) the id of responses:
• Disp = 1
• Vel = 2
• Accel = 3
• IncrDisp = 4
• IncrDeltaDisp = 5
• Reaction = 6
• Unbalance = 7
• RayleighForces = 8

1.6.22 nodeVel command

nodeVel(nodeTag, dof=-1)
Returns the current velocity at a specified node.

nodeTag node tag.

(int)
dof (int) specific dof at the node (1 through ndf), (optional), if no dof is provided, a list of values for
all dofs is returned.

1.6.23 nodeUnbalance command

nodeUnbalance(nodeTag, dof=-1)
Returns the unbalanced force at a specified node.

nodeTag node tag.

(int)
dof (int) specific dof at the node (1 through ndf), (optional), if no dof is provided, a list of values for
all dofs is returned.

1.6. Output Commands 261

OpenSeesPy Documentation, Release 3.5.1.3

1.6.24 numFact command

numFact()
Return the number of factorizations.

1.6.25 numIter command

numIter()
Return the number of iterations.

1.6.26 printA command

printA(’-file’, filename, ’-ret’)

print the contents of a FullGeneral system that the integrator creates to the screen or a file if the '-file'
option is used. If using a static integrator, the resulting matrix is the stiffness matrix. If a transient integrator,
it will be some combination of mass and stiffness matrices. The printA command can only be issued after an
analyze command.

filename (str) name of file to which output is sent, by default, print to the screen. (optional)
'-ret' (str) return the A matrix as a list. (optional)

1.6.27 printB command

printB(’-file’, filename, ’-ret’)

print the right hand side of a FullGeneral system that the integrator creates to the screen or a file if the '-file'
option is used.

filename (str) name of file to which output is sent, by default, print to the screen. (optional)
'-ret' (str) return the B vector as a list. (optional)

1.6.28 printGID command

printGID(filename, ’-append’, ’-eleRange’, startEle, endEle)

Print in GID format.

filename (str) output file name.

'-append' (str) append to existing file. (optional)
startEle (int) start element tag. (optional)
endEle (int) end element tag. (optional)

1.6.29 printModel command

printModel(’-JSON’, ’-file’, filename, ’-node’, ’-flag’, flag, *nodes=[], *eles=[])

This command is used to print output to screen or file.

262 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

filename name of file to which output is sent, by default, print to the screen. (optional)
(str)
'-JSON' (str) print to a JSON file. (optional)
'-node' (str) print node information. (optional)
flag (int) integer flag to be sent to the print() method, depending on the node and element type
(optional)
nodes (list a list of nodes tags to be printed, default is to print all, (optional)
(int))
eles (list (int)) a list of element tags to be printed, default is to print all, (optional)

Note: This command was called print in Tcl. Since print is a built-in function in Python, it is renamed to
printModel.

1.6.30 record command

record()
This command is used to cause all the recorders to do a record on the current state of the model.

Note: A record is issued after every successfull static or transient analysis step. Sometimes the user may need the
record to be issued on more occasions than this, for example if the user is just looking to record the eigenvectors after
an eigen command or for example the user wishes to include the state of the model at time 0.0 before any analysis has
been completed.

1.6.31 recorder command

recorder(recorderType, *recorderArgs)
This command is used to generate a recorder object which is to monitor what is happening during the analysis
and generate output for the user.
Return:
• >0 an integer tag that can be used as a handle on the recorder for the remove recorder commmand.
• -1 recorder command failed if integer -1 returned.

recorderType (str) recorder type

recorderArgs (list) a list of recorder arguments

The following contain information about available recorderType:

1. node recorder command
2. node envelope recorder command
3. element recorder command
4. element envelope recorder command
5. pvd recorder command
6. background recorder command

1.6. Output Commands 263

OpenSeesPy Documentation, Release 3.5.1.3

7. Collapse Recorder command

node recorder command

recorder(’Node’, ’-file’, filename, ’-xml’, filename, ’-binary’, filename, ’-tcp’, inetAddress, port, ’-
precision’, nSD=6, ’-timeSeries’, tsTag, ’-time’, ’-dT’, deltaT=0.0, ’-closeOnWrite’, ’-node’,
*nodeTags=[], ’-nodeRange’, startNode, endNode, ’-region’, regionTag, ’-dof’, *dofs=[], resp-
Type)
The Node recorder type records the response of a number of nodes at every converged step.

filename (str) name of file to which output is sent. file output

is either in xml format ('-xml' option), textual
('-file' option) or binary ('-binary' option)
which must pre-exist.
inetAddr (str) ip address, “xx.xx.xx.xx”, of remote machine to
which data is sent. (optional)
port (int) port on remote machine awaiting tcp. (optional)
nSD (int) number of significant digits (optional)
'-time' (str) using this option places domain time in first entry
of each data line, default is to have time ommitted,
(optional)
'-closeOnWrite' (str) using this option will instruct the recorder to invoke
a close on the data handler after every timestep. If
this is a file it will close the file on every step and
then re-open it for the next step. Note, this greatly
slows the execution time, but is useful if you need to
monitor the data during the analysis. (optional)
deltaT (float) time interval for recording. will record when next
step is deltaT greater than last recorder step. (op-
tional, default: records at every time step)
tsTag (int) the tag of a previously constructed TimeSeries, re-
sults from node at each time step are added to load
factor from series (optional)
nodeTags (list (int)) list of tags of nodes whose response is being
recorded (optional)
startNode (int) tag for start node whose response is being recorded
(optional)
endNode (int) tag for end node whose response is being recorded
(optional)
regionTag (int) a region tag; to specify all nodes in the previously
defined region. (optional)
dofs (list (int)) the specified dof at the nodes whose response is re-
quested.
resType (list (str)) a string indicating response required. Response
types are given in table below
• 'disp' displacement
• 'vel' velocity
• 'accel' acceleration
• 'incrDisp' incremental displacement
• 'reaction' nodal reaction
• 'eigen i' eigenvector for mode i
• 'rayleighForces' damping forces

264 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Note: Only one of '-file', '-xml', '-binary', '-tcp' will be used. If multiple specified last option is
used.

node envelope recorder command

recorder(’EnvelopeNode’, ’-file’, filename, ’-xml’, filename, ’-precision’, nSD=6, ’-timeSeries’, tsTag,

’-time’, ’-dT’, deltaT=0.0, ’-closeOnWrite’, ’-node’, *nodeTags=[], ’-nodeRange’, startNode,
endNode, ’-region’, regionTag, ’-dof’, *dofs=[], respType)
The EnvelopeNode recorder type records the min, max and absolute max of a number of nodal response quan-
taties.

filename (str) name of file to which output is sent. file output is

either in xml format ('-xml' option), or textual
('-file' option) which must pre-exist.
nSD (int) number of significant digits (optional)
'-time' (str) using this option places domain time in first entry
of each data line, default is to have time ommitted,
(optional)
'-closeOnWrite' (str) using this option will instruct the recorder to invoke
a close on the data handler after every timestep. If
this is a file it will close the file on every step and
then re-open it for the next step. Note, this greatly
slows the execution time, but is useful if you need to
monitor the data during the analysis. (optional)
deltaT (float) time interval for recording. will record when next
step is deltaT greater than last recorder step. (op-
tional, default: records at every time step)
tsTag (int) the tag of a previously constructed TimeSeries, re-
sults from node at each time step are added to load
factor from series (optional)
nodeTags (list (int)) list of tags of nodes whose response is being
recorded (optional)
startNode (int) tag for start node whose response is being recorded
(optional)
endNode (int) tag for end node whose response is being recorded
(optional)
regionTag (int) a region tag; to specify all nodes in the previously
defined region. (optional)
dofs (list (int)) the specified dof at the nodes whose response is re-
quested.
resType (list (str)) a string indicating response required. Response
types are given in table below
• 'disp' displacement
• 'vel' velocity
• 'accel' acceleration
• 'incrDisp' incremental displacement
• 'reaction' nodal reaction
• 'eigen i' eigenvector for mode i

1.6. Output Commands 265

OpenSeesPy Documentation, Release 3.5.1.3

element recorder command

recorder(’Element’, ’-file’, filename, ’-xml’, filename, ’-binary’, filename, ’-precision’, nSD=6, ’-

timeSeries’, tsTag, ’-time’, ’-dT’, deltaT=0.0, ’-closeOnWrite’, ’-ele’, *eleTags=[], ’-eleRange’,
startEle, endEle, ’-region’, regionTag, *args)
The Element recorder type records the response of a number of elements at every converged step. The response
recorded is element-dependent and also depends on the arguments which are passed to the setResponse() element
method.

filenamename of file to which output is sent. file output is either in xml format ('-xml' option), textual
(str) ('-file' option) or binary ('-binary' option) which must pre-exist.
nSD number of significant digits (optional)
(int)
'-time' using this option places domain time in first entry of each data line, default is to have time
(str) ommitted, (optional)
using this option will instruct the recorder to invoke a close on the data handler after every
'-closeOnWrite'
(str) timestep. If this is a file it will close the file on every step and then re-open it for the next step.
Note, this greatly slows the execution time, but is useful if you need to monitor the data during
the analysis. (optional)
deltaT time interval for recording. will record when next step is deltaT greater than last recorder step.
(float) (optional, default: records at every time step)
tsTag the tag of a previously constructed TimeSeries, results from node at each time step are added to
(int) load factor from series (optional)
eleTags list of tags of elements whose response is being recorded (optional)
(list
(int))
startEletag for start node whose response is being recorded (optional)
(int)
endEle tag for end node whose response is being recorded (optional)
(int)
a region tag; to specify all nodes in the previously defined region. (optional)
regionTag
(int)
args arguments which are passed to the setResponse() element method, all arguments must be in string
(list) format even for double and integer numbers because internally the setResponse() element method
only accepts strings.

Note: The setResponse() element method is dependent on the element type, and is described with the element()
Command.

element envelope recorder command

recorder(’EnvelopeElement’, ’-file’, filename, ’-xml’, filename, ’-binary’, filename, ’-precision’, nSD=6, ’-

timeSeries’, tsTag, ’-time’, ’-dT’, deltaT=0.0, ’-closeOnWrite’, ’-ele’, *eleTags=[], ’-eleRange’,
startEle, endEle, ’-region’, regionTag, *args)
The Envelope Element recorder type records the response of a number of elements at every converged step.
The response recorded is element-dependent and also depends on the arguments which are passed to the setRe-
sponse() element method. When the object is terminated, through the use of a wipe, exit, or remove the object
will output the min, max and absolute max values on 3 seperate lines of the output file for each quantity.

266 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

filenamename of file to which output is sent. file output is either in xml format ('-xml' option), textual
(str) ('-file' option) or binary ('-binary' option) which must pre-exist.
nSD number of significant digits (optional)
(int)
'-time' using this option places domain time in first entry of each data line, default is to have time
(str) ommitted, (optional)
using this option will instruct the recorder to invoke a close on the data handler after every
'-closeOnWrite'
(str) timestep. If this is a file it will close the file on every step and then re-open it for the next step.
Note, this greatly slows the execution time, but is useful if you need to monitor the data during
the analysis. (optional)
deltaT time interval for recording. will record when next step is deltaT greater than last recorder step.
(float) (optional, default: records at every time step)
tsTag the tag of a previously constructed TimeSeries, results from node at each time step are added to
(int) load factor from series (optional)
eleTags list of tags of elements whose response is being recorded (optional)
(list
(int))
startEletag for start node whose response is being recorded (optional)
(int)
endEle tag for end node whose response is being recorded (optional)
(int)
a region tag; to specify all nodes in the previously defined region. (optional)
regionTag
(int)
args arguments which are passed to the setResponse() element method
(list)

Note: The setResponse() element method is dependent on the element type, and is described with the element()
Command.

pvd recorder command

recorder(’PVD’, filename, ’-precision’, precision=10, ’-dT’, dT=0.0, *res)

Create a PVD recorder.

filename (str) the name for filename.pvd and filename/ di-

rectory, which must pre-exist.
precision (int) the precision of data. (optional)
dT (float) the time interval for recording. (optional)
res (list (str)) a list of (str) of responses to be recorded, (optional)
• 'disp'
• 'vel'
• 'accel'
• 'incrDisp'
• 'reaction'
• 'pressure'
• 'unbalancedLoad'
• 'mass'
• 'eigen'

1.6. Output Commands 267

OpenSeesPy Documentation, Release 3.5.1.3

background recorder command

recorder(’BgPVD’, filename, ’-precision’, precision=10, ’-dT’, dT=0.0, *res)

Create a PVD recorder for background mesh. This recorder is same as the PVD recorder, but will be automati-
cally called in background mesh and is able to record wave height and velocity.

filename (str) the name for filename.pvd and filename/ di-

rectory, which must pre-exist.
precision (int) the precision of data. (optional)
dT (float) the time interval for recording. (optional)
res (list (str)) a list of (str) of responses to be recorded, (optional)
• 'disp'
• 'vel'
• 'accel'
• 'incrDisp'
• 'reaction'
• 'pressure'
• 'unbalancedLoad'
• 'mass'
• 'eigen'

Collapse Recorder command

recorder(’Collapse’, ’-node’, nodeTag, ’-file_infill’, fileNameinf, ’-checknodes’, nTagbotn, nTagmidn, nTag-

topn, ’-global_gravaxis’, globgrav, ’-secondary’, ’-eles’, *eleTags, ’-eleRage’, start, end, ’-
region’, regionTag, ’-time’, ’-dT’, dT, ’-file’, fileName, ’-mass’, *massValues, ’-g’, gAcc, gDir,
gPat, ’-section’, *secTags, ’-crit’, critType, critValue)
A progressive collapse algorithm is developed by Talaat, M and Mosalam, K.M. in [1-3] and is implemented
in OpenSeesPy interpreter. The different applications of said algorithm are exemplified in references [4-7].
This algorithm is developed using element removal which relies on the dynamic equilibrium and the subsequent
transient change in the system kinematics. The theoretical background of the routine is detailed in the references
mentioned herein.

268 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

nodeTag (int) node tag

fileNameinf (str) is the file used to input the displacement interaction
curve. Two columns of data are input in this file
where only positive values are input. First column
is the OOP displacement in ascending order and sec-
ond column is the corresponding IP displacement.
Full interaction should be defined. In other words,
first value of OOP displacement and last value of IP
displacement should be zero.
fileName (str) is the file name for element removal log. Only one
log file is constructed for all collapse recorder com-
mands (i.e. for all removals). The first file name in-
put to a collapse recorder command is used and any
subsequent file names are ignored.
globgrav (float) is the global axis of the model in the direction of
gravity. 1, 2 and 3 should be input for X, Y and Z
axes, respectively.
critType (list (str)) criterial type
• 'INFILLWALL' no value required
• 'minStrain' value required
• 'maxStrain' value required
• 'axialDI' value required
• 'flexureDI' value required
• 'axialLS' value required
• 'shearLS' value required

The progressive collapse algorithm is thus implemented within OpenSeesPy for an automatic removal of elements
which have “numerically” collapse during an ongoing dynamic simulation. Main elements of the progressive collapse
routine are illustrated in Figures 1 and 2. The implementation is supported in Python as a relatively new OpenSees
module. Following each converged step of the dynamic analysis, the algorithm is called to check each element respec-
tively for possible violation given a user-defined removal criteria. The routine calls for the activation of the element
removal sequence before accessing the main analysis module on the subsequent analysis step. Activation of the el-
ement removal algorithm includes updating nodal masses, checking if the removal of the collapsed element results
in leaving behind dangling nodes or floating elements, which must be removed as well and removing all associated
element and nodal forces, imposed displacements, and constraints.

1.6. Output Commands 269

OpenSeesPy Documentation, Release 3.5.1.3

Furthermore, the aforementioned infill wall element and its removal criteria are defined for force- and
displacement-based distributed plasticity fiber elements and lumped plasticity beam–column elements with fiber-
discretized plastic hinges. Current version of OpenSeesPy considers only the removal of the infill wall
model described in (https://opensees.berkeley.edu/wiki/index.php/Infill_Wall_Model_and_Element_Removal#New_
Command_in_ OpenSees_Interpreter). Implementation of the removal of the elements representing the aforemen-
tioned infill wall analytical model in the progressive collapse algorithm is performed through defining a removal
criterion for the beam-column elements of this model. This criterion is based on the interaction between the in-plane
(IP) and out-of-plane (OOP) displacements. IP displacement is the relative horizontal displacement between the top
and bottom nodes of the diagonal element. OOP displacement is that of the middle node (where the OOP mass is
attached) with respectto the chord which connects the top and bottom nodes. The user is free to choose any interaction
relationship between IP and OOP displacements. In the example highlighted above, the interaction between in-plane
and out-of-plane is taken into consideration with regards to the displacement interaction between the two mechanisms,
where the IP and OOP displacement capacities are obtained using the FEMA 356 formulation for collapse prevention
level. During the nonlinear time history simulation, when the mentioned combination of displacements from the anal-
ysis exceeds the interaction curve, the two beam-column elements and the middle node, representing the unreinforced
masonry infill wall, are removed.
For the example illustrated in the next Figure, the existing Python command and its arguments in the OpenSeesPy
interpreter with respect to the infill wall removal is described such that:

recorder('Collapse', '-ele', ele1, '-time', '-crit', 'INFILLWALL', '-file', filename,

˓→'-file_infill', filenameinf, '-global_gravaxis, globgrav, '-checknodes', nodebot,

˓→nodemid, nodetop)

recorder('Collapse', '-ele', ele2, '-time', '-crit', 'INFILLWALL', '-file', filename,

˓→'-file_infill', filenameinf, '-global_gravaxis, globgrav, '-checknodes', nodebot,

˓→nodemid, nodetop)

recorder('Collapse', '-ele', ele1, ele2, '-node', nodemid)

270 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Note: it might seem that node inputs are unnecessary. However, when there are shear springs in the model, nodetop
and nodebot should be the nodes of the springs which connect to the beams, since the shear spring deformation
contributes to the IP displacement of the infill wall. These nodes are not the nodes of the diagonal element. Therefore,
it is necessary to input these nodes.

References
1. Talaat, M. and Mosalam, K.M. (2008), “Computational Modeling of Progressive Collapse in Reinforced Concrete
Frame Structures”, Pacific Earthquake Engineering Research Center, PEER 2007/10. 2. Talaat, M. and Mosalam,
K.M. (2009), “Modeling Progressive Collapse in Reinforced Concrete Buildings Using Direct Element Removal”,
Earthquake Engineering and Structural Dynamics, 38(5): 609-634. 3. Talaat, M. and Mosalam, K.M. (2009), Chap-
ter20: How to Simulate Column Collapse and Removal in As- built and Retrofitted Building Structures?, in Seismic
Risk Assessment and Retrofitting - with special emphasis on existing low-rise structures, Ilki, A, Karadogan, F, Pala,
S & Yuksel, E (Eds), ISBN 978-90-481-2680-4, Springer. 4. Talaat, M. and Mosalam, K.M. (2006), “Progressive
Collapse Modeling of Reinforced Concrete Framed Structures Containing Masonry Infill Walls”, Proceedings of the
2nd NEES/E-Defense Workshop on Collapse Simulation of Reinforced Concrete Building Structures, Kobe, Japan. 5.
Talaat, M. and Mosalam, K.M. (2007), “Towards Modeling Progressive Collapse in Reinforced Concrete Buildings”,
Proceedings of SEI-ASCE 2007 Structures Congress, Long Beach, California, USA. 6. Mosalam, K.M., Talaat, M.,
and Park, S. (2008), “Modeling Progressive Collapse in Reinforced Concrete Framed Structures”, Proceedings of the
14th World Conference on Earthquake Engineering, Beijing, China, October 12-17, Paper S15-018. 7. Mosalam,
K.M., Park, S., Günay, M.S. (2009), “Evaluation of an Element Removal Algorithm for Reinforced Concrete Struc-
tures Using Shake Table Experiments,” Proceedings of the 2nd International Conference on Computational Methods
in structural Dynamics and Earthquake Engineering (COMPDYN 2009), Island of Rhodes, Greece, June 22-24.

1.6.32 sectionForce command

sectionForce(eleTag, secNum, dof )

Returns the section force for a beam-column element. The dof of the section depends on the section type. Please
check with the section manual.

1.6. Output Commands 271

OpenSeesPy Documentation, Release 3.5.1.3

eleTag (int) element tag.

secNum (int) section number, i.e. the Gauss integratio number
dof (int) the dof of the section

1.6.33 sectionDeformation command

sectionDeformation(eleTag, secNum, dof )

Returns the section deformation for a beam-column element. The dof of the section depends on the section type.
Please check with the section manual.

eleTag (int) element tag.

secNum (int) section number, i.e. the Gauss integratio number
dof (int) the dof of the section

1.6.34 sectionStiffness command

sectionStiffness(eleTag, secNum)
Returns the section stiffness matrix for a beam-column element. A list of values in the row order will be returned.

eleTag (int) element tag.

secNum (int) section number, i.e. the Gauss integration number

1.6.35 sectionFlexibility command

sectionFlexibility(eleTag, secNum)
Returns the section flexibility matrix for a beam-column element. A list of values in the row order will be
returned.

eleTag (int) element tag.

secNum (int) section number, i.e. the Gauss integration number

1.6.36 sectionLocation command

sectionLocation(eleTag, secNum)
Returns the locations of integration points of a section for a beam-column element.

eleTag (int) element tag.

secNum (int) section number, i.e. the Gauss integration number

1.6.37 sectionWeight command

sectionWeight(eleTag, secNum)
Returns the weights of integration points of a section for a beam-column element.

eleTag (int) element tag.

secNum (int) section number, i.e. the Gauss integration number

272 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1.6.38 systemSize command

systemSize()
Return the size of the system.

1.6.39 testIter command

testIter()
Returns the number of iterations the convergence test took in the last analysis step

1.6.40 testNorm command

testNorm()
Returns the norms from the convergence test for the last analysis step.

Note: The size of norms will be equal to the max number of iterations specified. The first testIter of these will
be non-zero, the remaining ones will be zero.

1.6.41 version command

version()
Return the current OpenSees version.

1.6.42 logFile command

logFile(filename, ’-append’, ’-noEcho’)

Log all messages and errors in a file. By default, all messages and errors print to terminal or Jupyter Notebook
depending on how Python was run.

filename (str) name of the log file

'-append' (str) append to the file
'-noEcho' (str) do not print to terminal or Jupyter Notebook

1.7 Utility Commands

These commands are used to monitor and change the state of the model.
1. convertBinaryToText command
2. convertTextToBinary command
3. database command
4. InitialStateAnalysis command
5. loadConst command
6. modalDamping command
7. reactions command

1.7. Utility Commands 273

OpenSeesPy Documentation, Release 3.5.1.3

8. remove command
9. reset command
10. restore command
11. save command
12. sdfResponse command
13. setTime command
14. setNodeCoord command
15. setNodeDisp command
16. setNodeVel command
17. setNodeAccel command
18. setPrecision command
19. setElementRayleighDampingFactors command
20. start command
21. stop command
22. stripXML command
23. updateElementDomain command
24. updateMaterialStage
25. wipe command
26. wipeAnalysis command
27. setNumthread command
28. getNumthread command

1.7.1 convertBinaryToText command

convertBinaryToText(inputfile, outputfile)
Convert binary file to text file

inputfile (str) input file name.

outputfile (str) output file name.

1.7.2 convertTextToBinary command

convertTextToBinary(inputfile, outputfile)
Convert text file to binary file

inputfile (str) input file name.

outputfile (str) output file name.

274 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1.7.3 database command

database(type, dbName)
Create a database.

type (str) database type:

• 'File' - outputs database into a file
• 'MySQL' - creates a SQL database
• 'BerkeleyDB' - creates a BerkeleyDB
database

dbName (str) database name.

1.7.4 InitialStateAnalysis command

InitialStateAnalysis(flag)
Set the initial state analysis to 'on' or 'off'

flag (str) 'on' or 'off'

1.7.5 loadConst command

loadConst(’-time’, pseudoTime)
This command is used to set the loads constant in the domain and to also set the time in the domain. When
setting the loads constant, the procedure will invoke setLoadConst() on all LoadPattern objects which exist in
the domain at the time the command is called.

pseudoTime (float) Time domain is to be set to (optional)

Note: Load Patterns added afer this command is invoked are not set to constant.

1.7.6 modalDamping command

modalDamping(factor)
Set modal damping factor. The eigen() must be called before.

factor (float) damping factor.

1.7.7 reactions command

reactions(’-dynamic’, ’-rayleigh’)
Calculate the reactions. Call this command before the nodeReaction().

'-dynamic' (str) Include dynamic effects.

'-rayleigh' (str) Include rayleigh damping.

1.7. Utility Commands 275

OpenSeesPy Documentation, Release 3.5.1.3

1.7.8 remove command

remove(type, tag)
This commmand is used to remove components from the model.

type type of the object, 'ele', 'loadPattern', 'parameter', 'node', 'timeSeries',

(str) 'sp', 'mp'.
tag (int) tag of the object

remove(’recorders’)
Remove all recorder objects.
remove(’sp’, nodeTag, dofTag, patternTag)
Remove a sp object based on node

nodeTag (int) node tag

dof (int) dof the sp constrains
patternTag (int) pattern tag, (optional)

1.7.9 reset command

reset()
This command is used to set the state of the domain to its original state.

Note: It iterates over all components of the domain telling them to set their state back to the initial state. This is
not always the same as going back to the state of the model after initial model generation, e.g. if elements have been
removed.

1.7.10 restore command

restore(commitTag)
Restore data from database, which should be created through database().

commitTag (int) a tag identify the commit

1.7.11 save command

save(commitTag)
Save current state to database, which should be created through database().

commitTag (int) a tag identify the commit

1.7.12 sdfResponse command

sdfResponse(m, zeta, k, Fy, alpha, dtF, filename, dt[, uresidual, umaxprev ])

It is a command that computes bilinear single degree of freedom response in C++, and is much quicker than
using the OpenSees model builder. The command implements Newmark’s method with an inner Newton loop.

276 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

m (float) mass
zeta (float) damping ratio
k (float) stiffness
Fy (float) yielding strength
alpha (float) strain-hardening ratio
dtF (float) time step for input data
filename (str) input data file, one force per line
dt (float) time step for analysis
uresidual (float) residual displacement at the end of previous analysis (optional, default=0)
umaxprev (float) previous displacement (optional, default=0)

The command returns a list of five response quantities.

umax (float) maximum displacement during analysis

u (float) displacement at end of analysis
up (float) permanent residual displacement at end of analysis
amax (float) maximum acceleration during analysis
tamax (float) time when maximum accleration occurred

1.7.13 setTime command

setTime(pseudoTime)
This command is used to set the time in the Domain.

pseudoTime (float) Time domain to be set

1.7.14 setNodeCoord command

setNodeCoord(nodeTag, dim, value)

set the nodal coodinate at the specified dimension.

nodeTag (int) node tag.

dim (int) the dimension of the coordinate to be set.
value (float) coordinate value

1.7.15 setNodeDisp command

setNodeDisp(nodeTag, dof, value, ’-commit’)

set the nodal displacement at the specified DOF.

nodeTag (int) node tag.

dof (int) the DOF of the displacement to be set.
value (float) displacement value
'-commit' (str) commit nodal state. (optional)

1.7. Utility Commands 277

OpenSeesPy Documentation, Release 3.5.1.3

1.7.16 setNodeVel command

setNodeVel(nodeTag, dof, value, ’-commit’)

set the nodal velocity at the specified DOF.

nodeTag (int) node tag.

dof (int) the DOF of the velocity to be set.
value (float) velocity value
'-commit' (str) commit nodal state. (optional)

1.7.17 setNodeAccel command

setNodeAccel(nodeTag, dof, value, ’-commit’)

set the nodal acceleration at the specified DOF.

nodeTag (int) node tag.

dof (int) the DOF of the acceleration to be set.
value (float) acceleration value
'-commit' (str) commit nodal state. (optional)

1.7.18 setPrecision command

setPrecision(precision)
Set the precision for screen output.

precision (int) the precision number.

1.7.19 setElementRayleighDampingFactors command

setElementRayleighDampingFactors(eleTag, alphaM, betaK, betaK0, betaKc)

Set the rayleigh() damping for an element.

eleTag (int) element tag

alphaM (float) factor applied to elements or nodes mass matrix
betaK (float) factor applied to elements current stiffness matrix.
betaK0 (float) factor applied to elements initial stiffness matrix.
betaKc (float) factor applied to elements committed stiffness matrix.

1.7.20 start command

start()
Start the timer

1.7.21 stop command

stop()
Stop the timer and print timing information.

278 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1.7.22 stripXML command

stripXML(inputml, outputdata, outputxml)

Strip a xml file to a data file and a descriptive file.

inputxml (str) input xml file name.

outputdata (str) output data file name.
outputxml (str) output xml file name.

1.7.23 updateElementDomain command

updateElementDomain()
Update elements in the domain.

1.7.24 updateMaterialStage

updateMaterialStage(’-material’, matTag, ’-stage’, value, ’-parameter’, paramTag)

This function is used in geotechnical modeling to maintain elastic nDMaterial response during the application
of gravity loads. The material is then updated to allow for plastic strains during additional static loads or
earthquakes.

matTag (int) tag of nDMaterial

value (int) stage value
paramTag (int) tag of parameter (optional)

1.7.25 wipe command

wipe()
This command is used to destroy all constructed objects, i.e. all components of the model, all components of
the analysis and all recorders.
This command is used to start over without having to exit and restart the interpreter. It causes all elements,
nodes, constraints, loads to be removed from the domain. In addition it deletes all recorders, analysis objects
and all material objects created by the model builder.

1.7.26 wipeAnalysis command

wipeAnalysis()
This command is used to destroy all components of the Analysis object, i.e. any objects created with system,
numberer, constraints, integrator, algorithm, and analysis commands.

1.7.27 setNumthread command

setNumThreads(num)
set the number of threads to be used in the multi-threaded environment.

num (int) number of threades

1.7. Utility Commands 279

OpenSeesPy Documentation, Release 3.5.1.3

1.7.28 getNumthread command

getNumThreads()
return the total number of threads available

1.8 FSI Commands

These commands are related to the Fluid-Structure Interaction analysis in OpenSees.

1. mesh command
2. remesh command
3. PFEM integrator
4. PFEM SOE
5. PFEM test
6. PFEM analysis

1.8.1 mesh command

1. line mesh
2. triangular mesh
3. quad mesh
4. tetrahedron mesh
5. particle mesh
6. background mesh
mesh(type, tag, *args)
Create a mesh object. See below for available mesh types.

line mesh

mesh(’line’, tag, numnodes, *ndtags, id, ndf, meshsize, eleType=”, *eleArgs=[])

Create a line mesh object.

280 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

tag (int) mesh tag.

numnodes (int) number of nodes for defining consective lines.
ndtags (list (int)) the node tags
id (int) mesh id. Meshes with same id are considered as
same structure of fluid identity.
• id = 0 : not in FSI
• id > 0 : structure
• id < 0 : fluid

ndf (int) ndf for nodes to be created.

meshsize (float) mesh size.
eleType (str) the type of the element, (optional)
• Elastic Beam Column Element
– forceBeamColumn
– dispBeamColumn
if no type is given, only nodes are created

eleArgs (list) a list of element arguments. The arguments are same

as in the element commands, but without element
tag, and node tags. (optional)
For example,
eleArgs =
['elasticBeamColumn', A,
E, Iz, transfTag]

triangular mesh

mesh(’tri’, tag, numlines, *ltags, id, ndf, meshsize, eleType=”, *eleArgs=[])

Create a triangular mesh object.

1.8. FSI Commands 281

OpenSeesPy Documentation, Release 3.5.1.3

tag (int) mesh tag.

numlines (int) number of lines (line mesh) for defining a polygon.
ltags (list (int)) the line mesh tags
id (int) mesh id. Meshes with same id are considered as
same structure of fluid identity.
• id = 0 : not in FSI
• id > 0 : structure
• id < 0 : fluid

ndf (int) ndf for nodes to be created.

meshsize (float) mesh size.
eleType (str) the element type, (optional)
• PFEMElementBubble
• PFEMElementCompressible
• Tri31 Element
• Elastic Beam Column Element
• forceBeamColumn
• dispBeamColumn
if no type is given, only nodes are created. if beam
elements are given, beams are created instead of tri-
angular elements.
eleArgs (list) a list of element arguments. The arguments are same
as in the element commands, but without element
tag, and node tags. (optional)
For example,
eleArgs = ['PFEMElementBubble',
rho, mu, b1, b2, thickness, kappa]

quad mesh

mesh(’quad’, tag, numlines, *ltags, id, ndf, meshsize, eleType=”, *eleArgs=[])

Create a quad mesh object. The number of lines must be 4. These lines are continuous to form a loop.

282 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

tag (int) mesh tag.

numlines (int) number of lines (line mesh) for defining a polygon.
ltags (list (int)) the line mesh tags
id (int) mesh id. Meshes with same id are considered as
same structure of fluid identity.
• id = 0 : not in FSI
• id > 0 : structure
• id < 0 : fluid

ndf (int) ndf for nodes to be created.

meshsize (float) mesh size.
eleType (str) the element type, (optional)
• PFEMElementBubble
• PFEMElementCompressible
• Tri31 Element
• Elastic Beam Column Element
• forceBeamColumn
• dispBeamColumn
• Shell Element
if no type is given, only nodes are created. If beam
elements are given, beams are created instead of
quad elements. If triangular elements are given, they
are created by dividing one quad to two triangles.
eleArgs (list) a list of element arguments. The arguments are same
as in the element commands, but without element
tag, and node tags. (optional)
For example,
eleArgs = ['PFEMElementBubble',
rho, mu, b1, b2, thickness, kappa]

tetrahedron mesh

mesh(’tet’, tag, nummesh, *mtags, id, ndf, meshsize, eleType=”, *eleArgs=[])

Create a 3D tetrahedron mesh object.

tag (int) mesh tag.

nummesh (int) number of 2D mesh for defining a 3D body.
mtags (list (int)) the mesh tags
id (int) mesh id. Meshes with same id are considered as
same structure of fluid identity.
• id = 0 : not in FSI
• id > 0 : structure
• id < 0 : fluid

ndf (int) ndf for nodes to be created.

meshsize (float) mesh size.
eleType (str) the element type, (optional)
• FourNodeTetrahedron
if no type is given, only nodes are created.
eleArgs (list) a list of element arguments. The arguments are same
as in the element commands, but without element
tag, and node tags. (optional)

1.8. FSI Commands 283

OpenSeesPy Documentation, Release 3.5.1.3

particle mesh

mesh(’part’, tag, type, *pArgs, eleType=”, *eleArgs=[], ’-vel’, *vel0, ’-pressure’, p0)
Create or return a group of particles which will be used for background mesh.

tag (int) mesh tag.

type (str) type of the mesh
pArgs (list (float)) coordinates of points defining the mesh region nx,
ny, nz are number of particles in x, y, and z direc-
tions
• 'quad' : [x1, y1, x2, y2, x3, y3, x4, y4, nx,
ny]
Coordinates of four corners in counter-clock
wise order.
• 'cube' [[x1, y1, z1, x2, y2, z2, x3, y3, z3,
x4, y4,] z4, x5, y5, z5, x6, y6, z6, x7, y7,
z7, x8, y8, z8, nx, ny, nz]
Coordinates of four corners at bottom and at
top in counter-clock wise order
• 'tri' : [x1, y1, x2, y2, x3, y3, nx, ny]
Coordinates of three corners in counter-clock
wise order
• 'line' : [x1, y1, x2, y2, nx]
Coordinates of two ends in counter-clock wise
order
• 'pointlist' [[x1, y1, <z1>, vx1, vy1,
<vz1>,] ax1, ay1, <az1>, p1, x2, y2, <z2>,
vx2, vy2, <vz2>, ax2, ay2, <az2>, p2, ..]
input particles’ data in a list, in the order of co-
ordinates of last time step, current coordinates,
velocity, acceleration, and pressure.
• 'pointlist' without list
return a list of current particles’ data in this
mesh
[tag1, x1, y1, <z1>, vx1, vy1, <vz1>, ax1, ay1,
<az1>, p1, tag2, x2, y2, <z2>, vx2, vy2, <vz2>,
ax2, ay2, <az2>, p2, ..]
The format is similar to the input list, but with
an additional tag for each particle.

eleType (str) the element type, (optional)

• PFEMElementBubble
• PFEMElementCompressible
• Tri31 Element
if no type is given, only nodes are created

eleArgs (list) a list of element arguments. (optional, see line mesh

and triangular mesh)
vel0 (list (float)) a list of initial velocities. (optional)
p0 (float) initial pressure. (optional)

284 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

background mesh

mesh(’bg’, basicsize, *lower, *upper, ’-tol’, tol, ’-meshtol’, meshtol, ’-wave’, wavefilename, numl, *locations,
’-numsub’, numsub, ’-structure’, id, numnodes, *snodes, ’-largeSize’, level, *llower, *lupper)
Create a background mesh.

basicsize (float) basic mesh size

lower (list (float)) a list of coordinates of the lower point of the back-
ground region.
upper (list (float)) a list of coordinates of the uuper point of the back-
ground region.
tol (float) tolerance for intri check. (optional, default 1e-10)
meshtol (float) tolerance for cell boundary check. (optional, default
0.1)
wavefilename (str) a filename to record wave heights and velocities (op-
tional)
numl (int) number of locations to record wave (optional)
locations (list (float)) coordinates of the locations (optional)
id (int) structural id used to identify FSI and SSI (solid-
solid-interaction)
• id = 0 : ignore the structure
• id > 0 : both FSI and SSI
• id < 0 : only SSI

numsnodes (int) number of structural nodes (optional)

sNodes (list (int)) a list of structural nodes (optional)
level (int) some regions can have larger mesh size with larger
level. level = 1 means same as basic mesh
size.
llower (list (float)) a list of coordinates of the lower point of the region
with larger mesh size (optional)
lupper (list (float)) a list of coordinates of the upper point of the region
with larger mesh size(optional)

1.8.2 remesh command

remesh(alpha=-1.0)
• 𝛼 ≥ 0 for updating moving mesh.
• 𝛼 < 0 for updating background mesh.
If there are nodes shared by different mesh in the domain, the principles to decide what element should be used
for a triangle:
1. If all 3 nodes share the same mesh, use that mesh for the triangle.
2. If all 3 nodes share more than one mesh, use the mesh with eleArgs defined and lowest id.
3. If all 3 nodes are in different mesh, use the mesh with lowest id.
4. If the selected mesh id >= 0, skip the triangle.
5. If the selected mesh has no eleArgs, skip the triangle.

1.8. FSI Commands 285

OpenSeesPy Documentation, Release 3.5.1.3

alpha (float) Parameter for the 𝛼 method to construct a mesh from

the node cloud of moving meshes. (optional)
• 𝛼 = 0 : no elements are created
• large 𝛼 : all elements in the convex hull are
created
• 1.0 < 𝛼 < 2.0 : usually gives a good shape

1.8.3 PFEM integrator

integrator(’PFEM’)
Create a PFEM Integrator.

1.8.4 PFEM SOE

system(’PFEM’, ’-compressible’, ’-mumps’)

Create a incompressible PFEM system of equations using the Umfpack solver

-compressible Solve using a quasi-incompressible formulation. (optional)

-mumps Solve using the MUMPS solver. (optional, not supported on Windows)

1.8.5 PFEM test

test(’PFEM’, tolv, tolp, tolrv, tolrp, tolrelv, tolrelp, iter, maxincr, pFlag=0, nType=2)
Create a PFEM test, which check both increments and residual for velocities and pressures.

tolv (float) Tolerance for velocity increments

tolp (float) Tolerance for pressure increments
tolrv (float) Tolerance for velocity residual
tolrp (float) Tolerance for pressure residual
tolrv (float) Tolerance for relative velocity increments
tolrp (float) Tolerance for relative pressure increments
iter (int) Max number of iterations to check
maxincr (int) Max times for error increasing
pFlag (int) Print flag (optional):
• 0 print nothing.
• 1 print information on norms each time
test() is invoked.
• 2 print information on norms and number of it-
erations at end of successful test.
• 4 at each step it will print the norms and also
the ∆𝑈 and 𝑅(𝑈 ) vectors.
• 5 if it fails to converge at end of numIter it
will print an error message but return a suc-
cessfull test.

nType (int) Type of norm, (0 = max-norm, 1 = 1-norm, 2 = 2-

norm). (optional)

286 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1.8.6 PFEM analysis

analysis(’PFEM’, dtmax, dtmin, gravity, ratio=0.5)

Create a OpenSees PFEMAnalysis object.

dtmax (float) Maximum time steps.

dtmin (float) Mimimum time steps.
gravity (float) Gravity acceleration used to move isolated particles.
ratio (float) The ratio to reduce time steps if it was not converged. (optional)

1.9 Sensitivity Commands

These commands are for sensitivity analysis in OpenSees.

1. parameter command
2. addToParameter command
3. updateParameter command
4. setParameter command
5. getParamTags command
6. getParamValue command
7. computeGradients command
8. sensitivityAlgorithm command
9. sensNodeDisp command
10. sensNodeVel command
11. sensNodeAccel command
12. sensLambda command
13. sensSectionForce command
14. sensNodePressure command

1.9.1 parameter command

parameter(tag, <specific parameter args>)

In DDM-based FE response sensitivity analysis, the sensitivity parameters can be material, geometry or discrete
loading parameters.

tag (int) integer tag identifying the parameter.

<specific parameter depend on the object in the FE model encapsulating the desired param-
args> eters.

Note: Each parameter must be unique in the FE domain, and all parameter tags must be numbered sequentially
starting from 1.

1.9. Sensitivity Commands 287

OpenSeesPy Documentation, Release 3.5.1.3

Examples

1. To identify the elastic modulus, E, of the material 1 at section 3 of element 4, the <specific object arguments>
string becomes:

parameter(1, 'element', 4, 'section', 3, 'material', 1, 'E')

2. To identify the elastic modulus, E, of elastic section 3 of element 4 (for elastic section, no specific material need
to be defined), the <specific object arguments> string becomes:

parameter(1, 'element', 4, 'section', 3, 'E')

3. To parameterize E for element 4 with material 1 (no section need to be defined), the <specific object arguments>
string simplifies as:

parameter(1, 'element', 4, 'material', 1, 'E')

Note: Notice that the format of the <specific object arguments> is different for each considered ele-
ment/section/material. The specific set of parameters and the relative <specific object arguments> format will be
added in the future.

1.9.2 addToParameter command

addToParameter(tag, <specific parameter args>)

In case that more objects (e.g., element, section) are mapped to an existing parameter, the command can be used
to relate these additional objects to the specific parameter.

tag (int) integer tag identifying the parameter.

<specific parameter depend on the object in the FE model encapsulating the desired param-
args> eters.

1.9.3 updateParameter command

updateParameter(tag, newValue)
Once the parameters in FE model are defined, their value can be updated.

tag (int) integer tag identifying the parameter.

newValue (float) the updated value to which the parameter needs to be set.

Note: Scott M.H., Haukaas T. (2008). “Software framework for parameter updating and finite element response
sensitivity analysis.” Journal of Computing in Civil Engineering, 22(5):281-291.

1.9.4 setParameter command

setParameter(’-val’, newValue, <’-ele’, *eleTags>, <’-eleRange’, start, end>, <*args>)

set value for an element parameter

288 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

newValue (float) the updated value to which the parameter needs to be set.
eleTags (list (int)) a list of element tags
start (int) start element tag
end (int) end element tag
args (list (str)) a list of strings for the element parameter

1.9.5 getParamTags command

getParamTags()
Return a list of tags for all parameters

1.9.6 getParamValue command

getParamValue(tag)
Return the value of a parameter

tag (int) integer tag identifying the parameter.

1.9.7 computeGradients command

computeGradients()
This command is used to perform a sensitivity analysis. If the user wants to call this command, then the
'-computeByCommand' should be set in the sensitivityAlgorithm command.

1.9.8 sensitivityAlgorithm command

sensitivityAlgorithm(type)
This command is used to create a sensitivity algorithm.

type (str) the type of the sensitivity algorithm,

• '-computeAtEachStep' automatically
compute at the end of each step
• '-compuateByCommand' compute by
calling computeGradients.

1.9.9 sensNodeDisp command

sensNodeDisp(nodeTag, dof, paramTag)

Returns the current displacement sensitivity to a parameter at a specified node.

nodeTag (int) node tag

dof (int) specific dof at the node (1 through ndf)
paramTag (int) parameter tag

1.9. Sensitivity Commands 289

OpenSeesPy Documentation, Release 3.5.1.3

1.9.10 sensNodeVel command

sensNodeVel(nodeTag, dof, paramTag)

Returns the current velocity sensitivity to a parameter at a specified node.

nodeTag (int) node tag

dof (int) specific dof at the node (1 through ndf)
paramTag (int) parameter tag

1.9.11 sensNodeAccel command

sensNodeAccel(nodeTag, dof, paramTag)

Returns the current acceleration sensitivity to a parameter at a specified node.

nodeTag (int) node tag

dof (int) specific dof at the node (1 through ndf)
paramTag (int) parameter tag

1.9.12 sensLambda command

sensLambda(patternTag, paramTag)
Returns the current load factor sensitivity to a parameter in a load pattern.

patternTag (int) load pattern tag

paramTag (int) parameter tag

1.9.13 sensSectionForce command

sensSectionForce(eleTag, <secNum>, dof, paramTag)

Returns the current section force sensitivity to a parameter at a specified element and section.

eleTag (int) element tag

secNum (int) section number (optional)
dof (int) specific dof at the element (1 through element force ndf)
paramTag (int) parameter tag

1.9.14 sensNodePressure command

sensNodePressure(nodeTag, paramTag)
Returns the current pressure sensitivity to a parameter at a specified node.

nodeTag (int) node tag

paramTag (int) parameter tag

290 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1.10 Reliability Commands

These commands are for reliability analysis in OpenSees.

1. randomVariable command

1.10.1 randomVariable command

randomVariable(tag, dist, ’-mean’, mean, ’-stdv’, stdv, ’-startPoint’, startPoint, ’-parameters’, *params)
Create a random variable with user specified distribution

tag (int) random variable tag

dist (str) random variable distribution
• 'normal'
• 'lognormal'
• 'gamma'
• 'shiftedExponential'
• 'shiftedRayleigh'
• 'exponential'
• 'rayleigh'
• 'uniform'
• 'beta'
• 'type1LargestValue'
• 'type1SmallestValue'
• 'type2LargestValue'
• 'type3SmallestValue'
• 'chiSquare'
• 'gumbel'
• 'weibull'
• 'laplace'
• 'pareto'

mean (float) mean value

stdv (float) standard deviation
startPoint (float) starting point of the distribution
params (list (int)) a list of parameter tags

1.11 Parallel Commands

The parallel commands are currently only working in the Linux version. The parallel OpenSeesPy is similar to
OpenSeesMP, which requires users to divide the model to distributed processors.
You can still run the single-processor version as before. To run the parallel version, you have to install a MPI imple-
mentation, such as mpich. Then call your python scripts in the command line

mpiexec -np np python filename.py

where np is the number of processors to be used, python is the python interpreter, and filename.py is the script
name.
Inside the script, OpenSeesPy is still imported as

1.10. Reliability Commands 291

OpenSeesPy Documentation, Release 3.5.1.3

import openseespy.opensees as ops

Common problems:
1. Unmatch send/recv will cause deadlock.
2. Writing to the same files at the same from different processors will cause race conditions.
3. Poor model decomposition will cause load imbalance problem.
Following are commands related to parallel computing:
1. getPID command
2. getNP command
3. barrier command
4. send command
5. recv command
6. Bcast command
7. setStartNodeTag command
8. domainChange command
9. Parallel Plain Numberer
10. Parallel RCM Numberer
11. MUMPS Solver
12. Parallel DisplacementControl
13. partition command

1.11.1 getPID command

getPID()
Get the processor ID of the calling processor.

1.11.2 getNP command

getNP()
Get total number of processors.

1.11.3 barrier command

barrier()
Set a barrier for all processors, i.e., faster processors will pause here to wait for all processors to reach to this
point.

292 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1.11.4 send command

send(’-pid’, pid, *data)

Send information to another processor.

pid (int) ID of processor where data is sent to

data (list (int)) can be a list of integers
data (list (float)) can be a list of floats
data (str) can be a string

Note: send command and recv command must match and the order of calling both commands matters.

1.11.5 recv command

recv(’-pid’, pid)
Receive information from another processor.

pid (int) ID of processor where data is received from

pid (str) if pid is 'ANY', the processor can receive data from any processor.

Note: send command and recv command must match and the order of calling both commands matters.

1.11.6 Bcast command

Bcast(*data)
Broadcast information from processor 0 to all processors.

data (list (int)) can be a list of integers

data (list (float)) can be a list of floats
data (str) can be a string

Note: Run the same command to receive data sent from pid = 0.
For example,

if pid == 0:

data1 = []
data2 = []

ops.Bcast(*data1)
ops.Bcast(*data2)

if pid != 0:
data1 = ops.Bcast()
data2 = ops.Bcast()

1.11. Parallel Commands 293

OpenSeesPy Documentation, Release 3.5.1.3

1.11.7 setStartNodeTag command

setStartNodeTag(ndtag)
Set the starting node tag for the mesh command. The purpose of this command is to control the node tags
generated by the mesh command. Some nodes are shared by processors, which must have same tags. Nodes
which are unique to a processor must have uniques tags across all processors.

ndtag (int) starting node tag for the next call of mesh command

1.11.8 domainChange command

domainChange()
Mark the domain has changed manually. This is used to notify processors whose domain is not changed, but the
domain in other processors have changed.

1.11.9 partition command

partition(’-ncuts’, ncuts, ’-niter’, niters, ’-ufactor’, ufactor, ’-info’)

In a parallel environment, this command partitions the model. It requires that all processors have the exact same
model to be partitioned.

ncuts Specifies the number of different partitionings that it will compute. The final partitioning is the
(int) one that achieves the best edge cut or communication volume. (Optional default is 1).
niters Specifies the number of iterations for the refinement algorithms at each stage of the uncoarsening
(int) process. (Optional default is 10).
ufactor Specifies the maximum allowed load imbalance among the partitions. (Optional default is 30,
(int) indicating a load imbalance of 1.03).
'-info' print information. (optional)
(str)

1.12 Preprocessing Commands

The mesh command and remesh command should be called as

import openseespy.opensees as ops

ops.mesh()
ops.remesh()

The DiscretizeMember command should be called as

import openseespy.preprocessing.DiscretizeMember as opsdm

opsdm.DiscretizeMember()

1. mesh command
2. remesh command
3. DiscretizeMember command

294 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1.12.1 DiscretizeMember command

preprocessing.DiscretizeMember.DiscretizeMember(ndI, ndJ, numEle, eleType, integrTag,

transfTag, nodeTag, eleTag)
Discretize beam elements between two nodes.

ndI (int) node tag at I end

ndJ (int) node tag at J end
numEle (int) number of element to discretize
eleType (str) the element type
integrTag (int) beam integration tag (beamIntegration commands)
transfTag (int) geometric transformation tag (geomTransf commands)
nodeTag (int) starting node tag
eleTag (int) starting element tag

1.13 Postprocessing Modules

The postprocessing modules are removed from the OpenSeesPy module to their own modules for better maintainance
and support. Please direct any questions and issues to following documents for postprocessing modules:
1. opsvis
2. vfo (Visualization For OpenSees)

1.14 Examples

1. Structural Examples
2. Earthquake Examples
3. Tsunami Examples
4. GeoTechnical Examples
5. Thermal Examples
6. Parallel Examples
7. Plotting Examples
8. Sensitivity Examples

1.14.1 Structural Examples

1. Elastic Truss Analysis

2. Nonlinear Truss Analysis
3. Portal Frame 2d Analysis
4. Moment Curvature Analysis
5. Reinforced Concrete Frame Gravity Analysis
6. Reinforced Concrete Frame Pushover Analysis
7. Three story steel building with rigid beam-column connections and W-section

1.13. Postprocessing Modules 295

 OpenSeesPy Documentation, Release 3.5.1.3

8. Cantilever FRP-Confined Circular Reinforced Concrete Column under Cyclic Lateral Loading

Elastic Truss Analysis

1. The source code is shown below, which can be downloaded here.

2. Run the source code in your favorite Python program and should see Passed! in the results.

1 from openseespy.opensees import *

2

3 import numpy as np
4 import matplotlib.pyplot as plt
5

6 # ------------------------------
7 # Start of model generation
8 # -----------------------------
9

10 # remove existing model

11 wipe()
12

13 # set modelbuilder
14 model('basic', '-ndm', 2, '-ndf', 2)
15

16 # create nodes
17 node(1, 0.0, 0.0)
18 node(2, 144.0, 0.0)
19 node(3, 168.0, 0.0)
20 node(4, 72.0, 96.0)
21

22 # set boundary condition

23 fix(1, 1, 1)
24 fix(2, 1, 1)
25 fix(3, 1, 1)
26

27 # define materials
28 uniaxialMaterial("Elastic", 1, 3000.0)
29

30 # define elements
31 element("Truss",1,1,4,10.0,1)
32 element("Truss",2,2,4,5.0,1)
33 element("Truss",3,3,4,5.0,1)
34

35 # create TimeSeries
36 timeSeries("Linear", 1)
37

38 # create a plain load pattern

39 pattern("Plain", 1, 1)
40

41 # Create the nodal load - command: load nodeID xForce yForce

42 load(4, 100.0, -50.0)
43

44 # ------------------------------
45 # Start of analysis generation
46 # ------------------------------
47

48 # create SOE
49 system("BandSPD")
(continues on next page)

296 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

50

51 # create DOF number

52 numberer("RCM")
53

54 # create constraint handler

55 constraints("Plain")
56

57 # create integrator
58 integrator("LoadControl", 1.0)
59

60 # create algorithm
61 algorithm("Linear")
62

63 # create analysis object

64 analysis("Static")
65

66 # perform the analysis

67 analyze(1)
68

69 ux = nodeDisp(4,1)
70 uy = nodeDisp(4,2)
71 if abs(ux-0.53009277713228375450)<1e-12 and abs(uy+0.17789363846931768864)<1e-12:
72 print("Passed!")
73 else:
74 print("Failed!")

Nonlinear Truss Analysis

1. The source code is shown below, which can be downloaded here.

2. Make sure the numpy and matplotlib packages are installed in your Python distribution.
3. Run the source code in your favorite Python program and should see

1.14. Examples 297

 OpenSeesPy Documentation, Release 3.5.1.3

1 from openseespy.opensees import *

2

3 import numpy as np
4 import matplotlib.pyplot as plt
5

6 # ------------------------------
7 # Start of model generation
8 # -----------------------------
9

10 # set modelbuilder
11 wipe()
12 model('basic', '-ndm', 2, '-ndf', 2)
13

14 # variables
15 A = 4.0
16 E = 29000.0
17 alpha = 0.05
18 sY = 36.0
19 udisp = 2.5
20 Nsteps = 1000
21 Px = 160.0
22 Py = 0.0
23

24 # create nodes
25 node(1, 0.0, 0.0)
(continues on next page)

298 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

26 node(2, 72.0, 0.0)
27 node(3, 168.0, 0.0)
28 node(4, 48.0, 144.0)
29

30 # set boundary condition

31 fix(1, 1, 1)
32 fix(2, 1, 1)
33 fix(3, 1, 1)
34

35 # define materials
36 uniaxialMaterial("Hardening", 1, E, sY, 0.0, alpha/(1-alpha)*E)
37

38 # define elements
39 element("Truss",1,1,4,A,1)
40 element("Truss",2,2,4,A,1)
41 element("Truss",3,3,4,A,1)
42

43 # create TimeSeries
44 timeSeries("Linear", 1)
45

46 # create a plain load pattern

47 pattern("Plain", 1, 1)
48

49 # Create the nodal load

50 load(4, Px, Py)
51

52 # ------------------------------
53 # Start of analysis generation
54 # ------------------------------
55

56 # create SOE
57 system("ProfileSPD")
58

59 # create DOF number

60 numberer("Plain")
61

62 # create constraint handler

63 constraints("Plain")
64

65 # create integrator
66 integrator("LoadControl", 1.0/Nsteps)
67

68 # create algorithm
69 algorithm("Newton")
70

71 # create test
72 test('NormUnbalance',1e-8, 10)
73

74 # create analysis object

75 analysis("Static")
76

77 # ------------------------------
78 # Finally perform the analysis
79 # ------------------------------
80

81 # perform the analysis

82 data = np.zeros((Nsteps+1,2))
(continues on next page)

1.14. Examples 299

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

83 for j in range(Nsteps):
84 analyze(1)
85 data[j+1,0] = nodeDisp(4,1)
86 data[j+1,1] = getLoadFactor(1)*Px
87

88 plt.plot(data[:,0], data[:,1])
89 plt.xlabel('Horizontal Displacement')
90 plt.ylabel('Horizontal Load')
91 plt.show()
92

Portal Frame 2d Analysis

1. The source code is shown below, which can be downloaded here.

2. Run the source code in your favorite Python program and should see results below

Period Comparisons:
Period OpenSees SAP2000 SeismoStruct
1 1.27321 1.2732 1.2732
2 0.43128 0.4313 0.4313
3 0.24204 0.2420 0.2420
4 0.16018 0.1602 0.1602
5 0.11899 0.1190 0.1190
6 0.09506 0.0951 0.0951
7 0.07951 0.0795 0.0795

tSatic Analysis Result Comparisons:

Parameter OpenSees SAP2000 SeismoStruct
Disp Top 1.451 1.45 1.45
Axial Force Bottom Left 69.987 69.99 70.01
Moment Bottom Left 2324.677 2324.68 2324.71
PASSED Verification Test PortalFrame2d.py

1 from openseespy.opensees import *

2

3 from math import asin, sqrt

4

5 # Two dimensional Frame: Eigenvalue & Static Loads

6

8 # REFERENCES:
9 # used in verification by SAP2000:
10 # SAP2000 Integrated Finite Element Analysis and Design of Structures, Verification
˓→Manual,

11 # Computers and Structures, 1997. Example 1.

12 # and seismo-struct (Example 10)
13 # SeismoStruct, Verification Report For Version 6, 2012. Example 11.
14

15

16 # set some properties

17 wipe()
18

19 model('Basic', '-ndm', 2)
(continues on next page)

300 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

20

21 # properties
22

23 # units kip, ft
24

25 numBay = 2
26 numFloor = 7
27

28 bayWidth = 360.0
29 storyHeights = [162.0, 162.0, 156.0, 156.0, 156.0, 156.0, 156.0]
30

31 E = 29500.0
32 massX = 0.49
33 M = 0.
34 coordTransf = "Linear" # Linear, PDelta, Corotational
35 massType = "-lMass" # -lMass, -cMass
36

37 beams = ['W24X160', 'W24X160', 'W24X130', 'W24X130', 'W24X110', 'W24X110', 'W24X110']

38 eColumn = ['W14X246', 'W14X246', 'W14X246', 'W14X211', 'W14X211', 'W14X176', 'W14X176
˓→']

39 iColumn = ['W14X287', 'W14X287', 'W14X287', 'W14X246', 'W14X246', 'W14X211', 'W14X211

˓→']

40 columns = [eColumn, iColumn, eColumn]

41

42 WSection = {
43 'W14X176': [51.7, 2150.],
44 'W14X211': [62.1, 2670.],
45 'W14X246': [72.3, 3230.],
46 'W14X287': [84.4, 3910.],
47 'W24X110': [32.5, 3330.],
48 'W24X130': [38.3, 4020.],
49 'W24X160': [47.1, 5120.]
50 }
51

52 nodeTag = 1
53

54

55 # procedure to read
56 def ElasticBeamColumn(eleTag, iNode, jNode, sectType, E, transfTag, M, massType):
57 found = 0
58

59 prop = WSection[sectType]
60

61 A = prop[0]
62 I = prop[1]
63 element('elasticBeamColumn', eleTag, iNode, jNode, A, E, I, transfTag, '-mass', M,
˓→ massType)
64

65

66 # add the nodes

67 # - floor at a time
68 yLoc = 0.
69 for j in range(0, numFloor + 1):
70

71 xLoc = 0.
72 for i in range(0, numBay + 1):
73 node(nodeTag, xLoc, yLoc)
(continues on next page)

1.14. Examples 301

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

74 xLoc += bayWidth
75 nodeTag += 1
76

77 if j < numFloor:
78 storyHeight = storyHeights[j]
79

80 yLoc += storyHeight
81

82 # fix first floor

83 fix(1, 1, 1, 1)
84 fix(2, 1, 1, 1)
85 fix(3, 1, 1, 1)
86

87 # rigid floor constraint & masses

88 nodeTagR = 5
89 nodeTag = 4
90 for j in range(1, numFloor + 1):
91 for i in range(0, numBay + 1):
92

93 if nodeTag != nodeTagR:
94 equalDOF(nodeTagR, nodeTag, 1)
95 else:
96 mass(nodeTagR, massX, 1.0e-10, 1.0e-10)
97

98 nodeTag += 1
99

100 nodeTagR += numBay + 1

101

102 # add the columns

103 # add column element
104 geomTransf(coordTransf, 1)
105 eleTag = 1
106 for j in range(0, numBay + 1):
107

108 end1 = j + 1
109 end2 = end1 + numBay + 1
110 thisColumn = columns[j]
111

112 for i in range(0, numFloor):

113 secType = thisColumn[i]
114 ElasticBeamColumn(eleTag, end1, end2, secType, E, 1, M, massType)
115 end1 = end2
116 end2 += numBay + 1
117 eleTag += 1
118

119 # add beam elements

120 for j in range(1, numFloor + 1):
121 end1 = (numBay + 1) * j + 1
122 end2 = end1 + 1
123 secType = beams[j - 1]
124 for i in range(0, numBay):
125 ElasticBeamColumn(eleTag, end1, end2, secType, E, 1, M, massType)
126 end1 = end2
127 end2 = end1 + 1
128 eleTag += 1
129

130 # calculate eigenvalues & print results

(continues on next page)

302 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

131 numEigen = 7
132 eigenValues = eigen(numEigen)
133 PI = 2 * asin(1.0)
134

135 #
136 # apply loads for static analysis & perform analysis
137 #
138

139 timeSeries('Linear', 1)
140 pattern('Plain', 1, 1)
141 load(22, 20.0, 0., 0.)
142 load(19, 15.0, 0., 0.)
143 load(16, 12.5, 0., 0.)
144 load(13, 10.0, 0., 0.)
145 load(10, 7.5, 0., 0.)
146 load(7, 5.0, 0., 0.)
147 load(4, 2.5, 0., 0.)
148

149 integrator('LoadControl', 1.0)

150 algorithm('Linear')
151 analysis('Static')
152 analyze(1)
153

154 # determine PASS/FAILURE of test

155 ok = 0
156

157 #
158 # print pretty output of comparisons
159 #
160

161 # SAP2000 SeismoStruct

162 comparisonResults = [[1.2732, 0.4313, 0.2420, 0.1602, 0.1190, 0.0951, 0.0795],
163 [1.2732, 0.4313, 0.2420, 0.1602, 0.1190, 0.0951, 0.0795]]
164 print("\n\nPeriod Comparisons:")
165 print('{:>10}{:>15}{:>15}{:>15}'.format('Period', 'OpenSees', 'SAP2000', 'SeismoStruct
˓→'))

166

167 # formatString {%10s%15.5f%15.4f%15.4f}

168 for i in range(0, numEigen):
169 lamb = eigenValues[i]
170 period = 2 * PI / sqrt(lamb)
171 print('{:>10}{:>15.5f}{:>15.4f}{:>15.4f}'.format(i + 1, period,
˓→comparisonResults[0][i], comparisonResults[1][i]))

172 resultOther = comparisonResults[0][i]

173 if abs(period - resultOther) > 9.99e-5:
174 ok - 1
175

176 # print table of comparision

177 # Parameter SAP2000 SeismoStruct
178 comparisonResults = [["Disp Top", "Axial Force Bottom Left", "Moment Bottom Left"],
179 [1.45076, 69.99, 2324.68],
180 [1.451, 70.01, 2324.71]]
181 tolerances = [9.99e-6, 9.99e-3, 9.99e-3]
182

183 print("\n\nSatic Analysis Result Comparisons:")

184 print('{:>30}{:>15}{:>15}{:>15}'.format('Parameter', 'OpenSees', 'SAP2000',
˓→'SeismoStruct'))
(continues on next page)

1.14. Examples 303

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

185 for i in range(3):
186 response = eleResponse(1, 'forces')
187 if i == 0:
188 result = nodeDisp(22, 1)
189 elif i == 1:
190 result = abs(response[1])
191 else:
192 result = response[2]
193

194 print('{:>30}{:>15.3f}{:>15.2f}{:>15.2f}'.format(comparisonResults[0][i],
195 result,
196 comparisonResults[1][i],
197 comparisonResults[2][i]))
198 resultOther = comparisonResults[1][i]
199 tol = tolerances[i]
200 if abs(result - resultOther) > tol:
201 ok - 1
202 print("failed-> ", i, abs(result - resultOther), tol)
203

204 if ok == 0:
205 print("PASSED Verification Test PortalFrame2d.py \n\n")
206 else:
207 print("FAILED Verification Test PortalFrame2d.py \n\n")
208

Moment Curvature Analysis

1. The source code is shown below, which can be downloaded here.

2. Run the source code in your favorite Python program and should see results below

Start MomentCurvature.py example

Estimated yield curvature: 0.000126984126984127
Passed!
==========================

1 from openseespy.opensees import *

2

3 def MomentCurvature(secTag, axialLoad, maxK, numIncr=100):

4

5 # Define two nodes at (0,0)

6 node(1, 0.0, 0.0)
7 node(2, 0.0, 0.0)
8

9 # Fix all degrees of freedom except axial and bending

10 fix(1, 1, 1, 1)
11 fix(2, 0, 1, 0)
12

13 # Define element
14 # tag ndI ndJ secTag
15 element('zeroLengthSection', 1, 1, 2, secTag)
16

17 # Define constant axial load

18 timeSeries('Constant', 1)
19 pattern('Plain', 1, 1)
(continues on next page)

304 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

20 load(2, axialLoad, 0.0, 0.0)
21

22 # Define analysis parameters

23 integrator('LoadControl', 0.0)
24 system('SparseGeneral', '-piv')
25 test('NormUnbalance', 1e-9, 10)
26 numberer('Plain')
27 constraints('Plain')
28 algorithm('Newton')
29 analysis('Static')
30

31 # Do one analysis for constant axial load

32 analyze(1)
33

34 # Define reference moment

35 timeSeries('Linear', 2)
36 pattern('Plain',2, 2)
37 load(2, 0.0, 0.0, 1.0)
38

39 # Compute curvature increment

40 dK = maxK / numIncr
41

42 # Use displacement control at node 2 for section analysis

43 integrator('DisplacementControl', 2,3,dK,1,dK,dK)
44

45 # Do the section analysis

46 analyze(numIncr)
47

48

49 wipe()
50 print("Start MomentCurvature.py example")
51

52 # Define model builder

53 # --------------------
54 model('basic','-ndm',2,'-ndf',3)
55

56 # Define materials for nonlinear columns

57 # ------------------------------------------
58 # CONCRETE tag f'c ec0 f'cu ecu
59 # Core concrete (confined)
60 uniaxialMaterial('Concrete01',1, -6.0, -0.004, -5.0, -0.014)
61

62 # Cover concrete (unconfined)

63 uniaxialMaterial('Concrete01',2, -5.0, -0.002, 0.0, -0.006)
64

65 # STEEL
66 # Reinforcing steel
67 fy = 60.0 # Yield stress
68 E = 30000.0 # Young's modulus
69

70 # tag fy E0 b
71 uniaxialMaterial('Steel01', 3, fy, E, 0.01)
72

73 # Define cross-section for nonlinear columns

74 # ------------------------------------------
75

76 # set some paramaters

(continues on next page)

1.14. Examples 305

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

77 colWidth = 15
78 colDepth = 24
79

80 cover = 1.5
81 As = 0.60; # area of no. 7 bars
82

83 # some variables derived from the parameters

84 y1 = colDepth/2.0
85 z1 = colWidth/2.0
86

87

88 section('Fiber', 1)
89

90 # Create the concrete core fibers

91 patch('rect',1,10,1 ,cover-y1, cover-z1, y1-cover, z1-cover)
92

93 # Create the concrete cover fibers (top, bottom, left, right)

94 patch('rect',2,10,1 ,-y1, z1-cover, y1, z1)
95 patch('rect',2,10,1 ,-y1, -z1, y1, cover-z1)
96 patch('rect',2,2,1 ,-y1, cover-z1, cover-y1, z1-cover)
97 patch('rect',2,2,1 ,y1-cover, cover-z1, y1, z1-cover)
98

99 # Create the reinforcing fibers (left, middle, right)

100 layer('straight', 3, 3, As, y1-cover, z1-cover, y1-cover, cover-z1)
101 layer('straight', 3, 2, As, 0.0 , z1-cover, 0.0 , cover-z1)
102 layer('straight', 3, 3, As, cover-y1, z1-cover, cover-y1, cover-z1)
103

104 # Estimate yield curvature

105 # (Assuming no axial load and only top and bottom steel)
106 # d -- from cover to rebar
107 d = colDepth-cover
108 # steel yield strain
109 epsy = fy/E
110 Ky = epsy/(0.7*d)
111

112 # Print estimate to standard output

113 print("Estimated yield curvature: ", Ky)
114

115 # Set axial load

116 P = -180.0
117

118 # Target ductility for analysis

119 mu = 15.0
120

121 # Number of analysis increments

122 numIncr = 100
123

124 # Call the section analysis procedure

125 MomentCurvature(1, P, Ky*mu, numIncr)
126

127 results = open('results.out','a+')

128

129 u = nodeDisp(2,3)
130 if abs(u-0.00190476190476190541)<1e-12:
131 results.write('PASSED : MomentCurvature.py\n');
132 print("Passed!")
133 else:
(continues on next page)

306 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

134 results.write('FAILED : MomentCurvature.py\n');
135 print("Failed!")
136

137 results.close()
138

139 print("==========================")

Reinforced Concrete Frame Gravity Analysis

1. The source code is shown below, which can be downloaded here.

2. Run the source code in your favorite Python program and should see Passed! in the results.

1 print("==========================")
2

3 from openseespy.opensees import *

4

5 print("Starting RCFrameGravity example")

6

7 # Create ModelBuilder (with two-dimensions and 3 DOF/node)

8 model('basic', '-ndm', 2, '-ndf', 3)
9

10 # Create nodes
11 # ------------
12

13 # Set parameters for overall model geometry

14 width = 360.0
15 height = 144.0
16

17 # Create nodes
18 # tag, X, Y
19 node(1, 0.0, 0.0)
20 node(2, width, 0.0)
21 node(3, 0.0, height)
22 node(4, width, height)
23

24 # Fix supports at base of columns

25 # tag, DX, DY, RZ
26 fix(1, 1, 1, 1)
27 fix(2, 1, 1, 1)
28

29 # Define materials for nonlinear columns

30 # ------------------------------------------
31 # CONCRETE tag f'c ec0 f'cu ecu
32 # Core concrete (confined)
33 uniaxialMaterial('Concrete01', 1, -6.0, -0.004, -5.0, -0.014)
34

35 # Cover concrete (unconfined)

36 uniaxialMaterial('Concrete01', 2, -5.0, -0.002, 0.0, -0.006)
37

38 # STEEL
39 # Reinforcing steel
40 fy = 60.0; # Yield stress
41 E = 30000.0; # Young's modulus
42 # tag fy E0 b
(continues on next page)

1.14. Examples 307

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

43 uniaxialMaterial('Steel01', 3, fy, E, 0.01)
44

45 # Define cross-section for nonlinear columns

46 # ------------------------------------------
47

48 # some parameters
49 colWidth = 15
50 colDepth = 24
51

52 cover = 1.5
53 As = 0.60 # area of no. 7 bars
54

55 # some variables derived from the parameters

56 y1 = colDepth / 2.0
57 z1 = colWidth / 2.0
58

59 section('Fiber', 1)
60

61 # Create the concrete core fibers

62 patch('rect', 1, 10, 1, cover - y1, cover - z1, y1 - cover, z1 - cover)
63

64 # Create the concrete cover fibers (top, bottom, left, right)

65 patch('rect', 2, 10, 1, -y1, z1 - cover, y1, z1)
66 patch('rect', 2, 10, 1, -y1, -z1, y1, cover - z1)
67 patch('rect', 2, 2, 1, -y1, cover - z1, cover - y1, z1 - cover)
68 patch('rect', 2, 2, 1, y1 - cover, cover - z1, y1, z1 - cover)
69

70 # Create the reinforcing fibers (left, middle, right)

71 layer('straight', 3, 3, As, y1 - cover, z1 - cover, y1 - cover, cover - z1)
72 layer('straight', 3, 2, As, 0.0, z1 - cover, 0.0, cover - z1)
73 layer('straight', 3, 3, As, cover - y1, z1 - cover, cover - y1, cover - z1)
74

75 # Define column elements

76 # ----------------------
77

78 # Geometry of column elements

79 # tag
80

81 geomTransf('PDelta', 1)
82

83 # Number of integration points along length of element

84 np = 5
85

86 # Lobatto integratoin
87 beamIntegration('Lobatto', 1, 1, np)
88

89 # Create the coulumns using Beam-column elements

90 # e tag ndI ndJ transfTag integrationTag
91 eleType = 'forceBeamColumn'
92 element(eleType, 1, 1, 3, 1, 1)
93 element(eleType, 2, 2, 4, 1, 1)
94

95 # Define beam elment

96 # -----------------------------
97

98 # Geometry of column elements

99 # tag
(continues on next page)

308 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

100 geomTransf('Linear', 2)
101

102 # Create the beam element

103 # tag, ndI, ndJ, A, E, Iz, transfTag
104 element('elasticBeamColumn', 3, 3, 4, 360.0, 4030.0, 8640.0, 2)
105

106 # Define gravity loads

107 # --------------------
108

109 # a parameter for the axial load

110 P = 180.0; # 10% of axial capacity of columns
111

112 # Create a Plain load pattern with a Linear TimeSeries

113 timeSeries('Linear', 1)
114 pattern('Plain', 1, 1)
115

116 # Create nodal loads at nodes 3 & 4

117 # nd FX, FY, MZ
118 load(3, 0.0, -P, 0.0)
119 load(4, 0.0, -P, 0.0)
120

121 # ------------------------------
122 # End of model generation
123 # ------------------------------
124

125

126 # ------------------------------
127 # Start of analysis generation
128 # ------------------------------
129

130 # Create the system of equation, a sparse solver with partial pivoting
131 system('BandGeneral')
132

133 # Create the constraint handler, the transformation method

134 constraints('Transformation')
135

136 # Create the DOF numberer, the reverse Cuthill-McKee algorithm

137 numberer('RCM')
138

139 # Create the convergence test, the norm of the residual with a tolerance of
140 # 1e-12 and a max number of iterations of 10
141 test('NormDispIncr', 1.0e-12, 10, 3)
142

143 # Create the solution algorithm, a Newton-Raphson algorithm

144 algorithm('Newton')
145

146 # Create the integration scheme, the LoadControl scheme using steps of 0.1
147 integrator('LoadControl', 0.1)
148

149 # Create the analysis object

150 analysis('Static')
151

152 # ------------------------------
153 # End of analysis generation
154 # ------------------------------
155

156
(continues on next page)

1.14. Examples 309

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

157 # ------------------------------
158 # Finally perform the analysis
159 # ------------------------------
160

161 # perform the gravity load analysis, requires 10 steps to reach the load level
162 analyze(10)
163

164 # Print out the state of nodes 3 and 4

165 # print node 3 4
166

167 # Print out the state of element 1

168 # print ele 1
169

170 u3 = nodeDisp(3, 2)
171 u4 = nodeDisp(4, 2)
172

173 results = open('results.out', 'a+')

174

175 if abs(u3 + 0.0183736) < 1e-6 and abs(u4 + 0.0183736) < 1e-6:
176 results.write('PASSED : RCFrameGravity.py\n')
177 print("Passed!")
178 else:
179 results.write('FAILED : RCFrameGravity.py\n')
180 print("Failed!")
181

182 results.close()
183

184 print("==========================")

Reinforced Concrete Frame Pushover Analysis

1. The source code is shown below, which can be downloaded here.

2. The file for gravity analysis is also needed :here.
3. Run the source code in your favorite Python program and should see Passed! in the results.
1 print("==========================")
2 print("Start RCFramePushover Example")
3

4 # Units: kips, in, sec

5 #
6 # Written: GLF/MHS/fmk
7 # Date: January 2001
8 from openseespy.opensees import *
9

10 wipe()
11 # ----------------------------------------------------
12 # Start of Model Generation & Initial Gravity Analysis
13 # ----------------------------------------------------
14

15 # Do operations of Example3.1 by sourcing in the tcl file

16 import RCFrameGravity
17 print("Gravity Analysis Completed")
18

19 # Set the gravity loads to be constant & reset the time in the domain
(continues on next page)

310 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

20 loadConst('-time', 0.0)
21

22 # ----------------------------------------------------
23 # End of Model Generation & Initial Gravity Analysis
24 # ----------------------------------------------------
25

26

27 # ----------------------------------------------------
28 # Start of additional modelling for lateral loads
29 # ----------------------------------------------------
30

31 # Define lateral loads

32 # --------------------
33

34 # Set some parameters

35 H = 10.0 # Reference lateral load
36

37 # Set lateral load pattern with a Linear TimeSeries

38 pattern('Plain', 2, 1)
39

40 # Create nodal loads at nodes 3 & 4

41 # nd FX FY MZ
42 load(3, H, 0.0, 0.0)
43 load(4, H, 0.0, 0.0)
44

45 # ----------------------------------------------------
46 # End of additional modelling for lateral loads
47 # ----------------------------------------------------
48

49

50 # ----------------------------------------------------
51 # Start of modifications to analysis for push over
52 # ----------------------------------------------------
53

54 # Set some parameters

55 dU = 0.1 # Displacement increment
56

57 # Change the integration scheme to be displacement control

58 # node dof init Jd min max
59 integrator('DisplacementControl', 3, 1, dU, 1, dU, dU)
60

61 # ----------------------------------------------------
62 # End of modifications to analysis for push over
63 # ----------------------------------------------------
64

65

66 # ------------------------------
67 # Start of recorder generation
68 # ------------------------------
69

70 # Stop the old recorders by destroying them

71 # remove recorders
72

73 # Create a recorder to monitor nodal displacements

74 # recorder Node -file node32.out -time -node 3 4 -dof 1 2 3 disp
75

76 # Create a recorder to monitor element forces in columns

(continues on next page)

1.14. Examples 311

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

77 # recorder EnvelopeElement -file ele32.out -time -ele 1 2 forces
78

79 # --------------------------------
80 # End of recorder generation
81 # ---------------------------------
82

83

84 # ------------------------------
85 # Finally perform the analysis
86 # ------------------------------
87

88 # Set some parameters

89 maxU = 15.0 # Max displacement
90 currentDisp = 0.0
91 ok = 0
92

93 test('NormDispIncr', 1.0e-12, 1000)

94 algorithm('ModifiedNewton', '-initial')
95

96 while ok == 0 and currentDisp < maxU:

97

98 ok = analyze(1)
99

100 # if the analysis fails try initial tangent iteration

101 if ok != 0:
102 print("modified newton failed")
103 break
104 # print "regular newton failed .. lets try an initail stiffness for this step"
105 # test('NormDispIncr', 1.0e-12, 1000)
106 # # algorithm('ModifiedNewton', '-initial')
107 # ok = analyze(1)
108 # if ok == 0:
109 # print "that worked .. back to regular newton"
110

111 # test('NormDispIncr', 1.0e-12, 10)

112 # algorithm('Newton')
113

114 currentDisp = nodeDisp(3, 1)

115

116 results = open('results.out', 'a+')

117

118 if ok == 0:
119 results.write('PASSED : RCFramePushover.py\n')
120 print("Passed!")
121 else:
122 results.write('FAILED : RCFramePushover.py\n')
123 print("Failed!")
124

125 results.close()
126

127 # Print the state at node 3

128 # print node 3
129

130

131 print("==========================")

312 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Three story steel building with rigid beam-column connections and W-section

1. The source code is developed by Anurag Upadhyay from University of Utah.

2. The source code is shown below, which can be downloaded here.
3. Run the source code in your favorite Python program and should see following plot.

2 ##################################################################
3 ## 2D steel frame example.
4 ## 3 story steel building with rigid beam-column connections.
5 ## This script uses W-section command inOpensees to create steel..
6 ## .. beam-column fiber sections.
7 ##
8 ## By - Anurag Upadhyay, PhD Student, University of Utah.
9 ## Date - 08/06/2018
10 ##################################################################
11

12 print("=========================================================")
13 print("Start 2D Steel Frame Example")
14

15 from openseespy.opensees import *

16

17 import numpy as np
18 import matplotlib.pyplot as plt
(continues on next page)

1.14. Examples 313

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

19 import os
20

21 AnalysisType='Pushover' ; # Pushover Gravity

22

23 ## ------------------------------
24 ## Start of model generation
25 ## -----------------------------
26 # remove existing model
27 wipe()
28

29 # set modelbuilder
30 model('basic', '-ndm', 2, '-ndf', 3)
31

32 import math
33

34 ############################################
35 ### Units and Constants ###################
36 ############################################
37

38 inch = 1;
39 kip = 1;
40 sec = 1;
41

42 # Dependent units
43 sq_in = inch*inch;
44 ksi = kip/sq_in;
45 ft = 12*inch;
46

47 # Constants
48 g = 386.2*inch/(sec*sec);
49 pi = math.acos(-1);
50

51 #######################################
52 ##### Dimensions
53 #######################################
54

55 # Dimensions Input
56 H_story=10.0*ft;
57 W_bayX=16.0*ft;
58 W_bayY_ab=5.0*ft+10.0*inch;
59 W_bayY_bc=8.0*ft+4.0*inch;
60 W_bayY_cd=5.0*ft+10.0*inch;
61

62 # Calculated dimensions
63 W_structure=W_bayY_ab+W_bayY_bc+W_bayY_cd;
64

65 ################
66 ### Material
67 ################
68

69 # Steel02 Material
70

71 matTag=1;
72 matConnAx=2;
73 matConnRot=3;
74

75 Fy=60.0*ksi; # Yield stress

(continues on next page)

314 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

76 Es=29000.0*ksi; # Modulus of Elasticity of Steel
77 v=0.2; # Poisson's ratio
78 Gs=Es/(1+v); # Shear modulus
79 b=0.10; # Strain hardening ratio
80 params=[18.0,0.925,0.15] # R0,cR1,cR2
81 R0=18.0
82 cR1=0.925
83 cR2=0.15
84 a1=0.05
85 a2=1.00
86 a3=0.05
87 a4=1.0
88 sigInit=0.0
89 alpha=0.05
90

91 uniaxialMaterial('Steel02', matTag, Fy, Es, b, R0, cR1, cR2, a1, a2, a3, a4, sigInit)
92

93 # ##################
94 # ## Sections
95 # ##################
96

97 colSecTag1=1;
98 colSecTag2=2;
99 beamSecTag1=3;
100 beamSecTag2=4;
101 beamSecTag3=5;
102

103 # COMMAND: section('WFSection2d', secTag, matTag, d, tw, bf, tf, Nfw, Nff)
104

105 section('WFSection2d', colSecTag1, matTag, 10.5*inch, 0.26*inch, 5.77*inch, 0.44*inch,

˓→ 15, 16) # outer Column
106 section('WFSection2d', colSecTag2, matTag, 10.5*inch, 0.26*inch, 5.77*inch, 0.44*inch,
˓→ 15, 16) # Inner Column
107

108 section('WFSection2d', beamSecTag1, matTag, 8.3*inch, 0.44*inch, 8.11*inch, 0.

˓→685*inch, 15, 15) # outer Beam
109 section('WFSection2d', beamSecTag2, matTag, 8.2*inch, 0.40*inch, 8.01*inch, 0.
˓→650*inch, 15, 15) # Inner Beam
110 section('WFSection2d', beamSecTag3, matTag, 8.0*inch, 0.40*inch, 7.89*inch, 0.
˓→600*inch, 15, 15) # Inner Beam
111

112 # Beam size - W10x26

113 Abeam=7.61*inch*inch;
114 IbeamY=144.*(inch**4); # Inertia along horizontal axis
115 IbeamZ=14.1*(inch**4); # inertia along vertical axis
116

117 # BRB input data

118 Acore=2.25*inch;
119 Aend=10.0*inch;
120 LR_BRB=0.55;
121

122 # ###########################
123 # ##### Nodes
124 # ###########################
125

126 # Create All main nodes

127 node(1, 0.0, 0.0)
(continues on next page)

1.14. Examples 315

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

128 node(2, W_bayX, 0.0)
129 node(3, 2*W_bayX, 0.0)
130

131 node(11, 0.0, H_story)

132 node(12, W_bayX, H_story)
133 node(13, 2*W_bayX, H_story)
134

135 node(21, 0.0, 2*H_story)

136 node(22, W_bayX, 2*H_story)
137 node(23, 2*W_bayX, 2*H_story)
138

139 node(31, 0.0, 3*H_story)

140 node(32, W_bayX, 3*H_story)
141 node(33, 2*W_bayX, 3*H_story)
142

143 # Beam Connection nodes

144

145 node(1101, 0.0, H_story)

146 node(1201, W_bayX, H_story)
147 node(1202, W_bayX, H_story)
148 node(1301, 2*W_bayX, H_story)
149

150 node(2101, 0.0, 2*H_story)

151 node(2201, W_bayX, 2*H_story)
152 node(2202, W_bayX, 2*H_story)
153 node(2301, 2*W_bayX, 2*H_story)
154

155 node(3101, 0.0, 3*H_story)

156 node(3201, W_bayX, 3*H_story)
157 node(3202, W_bayX, 3*H_story)
158 node(3301, 2*W_bayX, 3*H_story)
159

160 # ###############
161 # Constraints
162 # ###############
163

164 fix(1, 1, 1, 1)
165 fix(2, 1, 1, 1)
166 fix(3, 1, 1, 1)
167

168 # #######################
169 # ### Elements
170 # #######################
171

172 # ### Assign beam-integration tags

173

174 ColIntTag1=1;
175 ColIntTag2=2;
176 BeamIntTag1=3;
177 BeamIntTag2=4;
178 BeamIntTag3=5;
179

180 beamIntegration('Lobatto', ColIntTag1, colSecTag1, 4)

181 beamIntegration('Lobatto', ColIntTag2, colSecTag2, 4)
182 beamIntegration('Lobatto', BeamIntTag1, beamSecTag1, 4)
183 beamIntegration('Lobatto', BeamIntTag2, beamSecTag2, 4)
184 beamIntegration('Lobatto', BeamIntTag3, beamSecTag3, 4)
(continues on next page)

316 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

185

186 # Assign geometric transformation

187

188 ColTransfTag=1
189 BeamTranfTag=2
190

191 geomTransf('PDelta', ColTransfTag)

192 geomTransf('Linear', BeamTranfTag)
193

194

195 # Assign Elements ##############

196

197 # ## Add non-linear column elements

198 element('forceBeamColumn', 1, 1, 11, ColTransfTag, ColIntTag1, '-mass', 0.0)
199 element('forceBeamColumn', 2, 2, 12, ColTransfTag, ColIntTag2, '-mass', 0.0)
200 element('forceBeamColumn', 3, 3, 13, ColTransfTag, ColIntTag1, '-mass', 0.0)
201

202 element('forceBeamColumn', 11, 11, 21, ColTransfTag, ColIntTag1, '-mass', 0.0)

203 element('forceBeamColumn', 12, 12, 22, ColTransfTag, ColIntTag2, '-mass', 0.0)
204 element('forceBeamColumn', 13, 13, 23, ColTransfTag, ColIntTag1, '-mass', 0.0)
205

206 element('forceBeamColumn', 21, 21, 31, ColTransfTag, ColIntTag1, '-mass', 0.0)

207 element('forceBeamColumn', 22, 22, 32, ColTransfTag, ColIntTag2, '-mass', 0.0)
208 element('forceBeamColumn', 23, 23, 33, ColTransfTag, ColIntTag1, '-mass', 0.0)
209

210 #
211

212 # ### Add linear main beam elements, along x-axis

213 #element('elasticBeamColumn', 101, 1101, 1201, Abeam, Es, Gs, Jbeam, IbeamY, IbeamZ,
˓→beamTransfTag, '-mass', 0.0)

214

215 element('forceBeamColumn', 101, 1101, 1201, BeamTranfTag, BeamIntTag1, '-mass', 0.0)

216 element('forceBeamColumn', 102, 1202, 1301, BeamTranfTag, BeamIntTag1, '-mass', 0.0)
217

218 element('forceBeamColumn', 201, 2101, 2201, BeamTranfTag, BeamIntTag2, '-mass', 0.0)

219 element('forceBeamColumn', 202, 2202, 2301, BeamTranfTag, BeamIntTag2, '-mass', 0.0)
220

221 element('forceBeamColumn', 301, 3101, 3201, BeamTranfTag, BeamIntTag3, '-mass', 0.0)

222 element('forceBeamColumn', 302, 3202, 3301, BeamTranfTag, BeamIntTag3, '-mass', 0.0)
223

224 # Assign constraints between beam end nodes and column nodes (RIgid beam column
˓→connections)

225 equalDOF(11, 1101, 1,2,3)

226 equalDOF(12, 1201, 1,2,3)
227 equalDOF(12, 1202, 1,2,3)
228 equalDOF(13, 1301, 1,2,3)
229

230 equalDOF(21, 2101, 1,2,3)

231 equalDOF(22, 2201, 1,2,3)
232 equalDOF(22, 2202, 1,2,3)
233 equalDOF(23, 2301, 1,2,3)
234

235 equalDOF(31, 3101, 1,2,3)

236 equalDOF(32, 3201, 1,2,3)
237 equalDOF(32, 3202, 1,2,3)
238 equalDOF(33, 3301, 1,2,3)
239
(continues on next page)

1.14. Examples 317

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

240

241 ################
242 ## Gravity Load
243 ################
244 # create TimeSeries
245 timeSeries("Linear", 1)
246

247 # create a plain load pattern

248 pattern("Plain", 1, 1)
249

250 # Create the nodal load

251 load(11, 0.0, -5.0*kip, 0.0)
252 load(12, 0.0, -6.0*kip, 0.0)
253 load(13, 0.0, -5.0*kip, 0.0)
254

255 load(21, 0., -5.*kip, 0.0)

256 load(22, 0., -6.*kip,0.0)
257 load(23, 0., -5.*kip, 0.0)
258

259 load(31, 0., -5.*kip, 0.0)

260 load(32, 0., -6.*kip, 0.0)
261 load(33, 0., -5.*kip, 0.0)
262

263

264 # ------------------------------
265 # Start of analysis generation
266 # ------------------------------
267

268 NstepsGrav = 10
269

270 system("BandGEN")
271 numberer("Plain")
272 constraints("Plain")
273 integrator("LoadControl", 1.0/NstepsGrav)
274 algorithm("Newton")
275 test('NormUnbalance',1e-8, 10)
276 analysis("Static")
277

278

279 # perform the analysis

280 data = np.zeros((NstepsGrav+1,2))
281 for j in range(NstepsGrav):
282 analyze(1)
283 data[j+1,0] = nodeDisp(31,2)
284 data[j+1,1] = getLoadFactor(1)*5
285

286 loadConst('-time', 0.0)

287

288 print("Gravity analysis complete")

289

290 wipeAnalysis()
291

292 ###############################
293 ### PUSHOVER ANALYSIS
294 ###############################
295

296 if(AnalysisType=="Pushover"):
(continues on next page)

318 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

297

298 print("<<<< Running Pushover Analysis >>>>")

299

300 # Create load pattern for pushover analysis

301 # create a plain load pattern
302 pattern("Plain", 2, 1)
303

304 load(11, 1.61, 0.0, 0.0)

305 load(21, 3.22, 0.0, 0.0)
306 load(31, 4.83, 0.0, 0.0)
307

308 ControlNode=31
309 ControlDOF=1
310 MaxDisp=0.15*H_story
311 DispIncr=0.1
312 NstepsPush=int(MaxDisp/DispIncr)
313

314 system("ProfileSPD")
315 numberer("Plain")
316 constraints("Plain")
317 integrator("DisplacementControl", ControlNode, ControlDOF, DispIncr)
318 algorithm("Newton")
319 test('NormUnbalance',1e-8, 10)
320 analysis("Static")
321

322 PushDataDir = r'PushoverOut'

323 if not os.path.exists(PushDataDir):
324 os.makedirs(PushDataDir)
325 recorder('Node', '-file', "PushoverOut/Node2React.out", '-closeOnWrite', '-
˓→node', 2, '-dof',1, 'reaction')

326 recorder('Node', '-file', "PushoverOut/Node31Disp.out", '-closeOnWrite', '-

˓→node', 31, '-dof',1, 'disp')

327 recorder('Element', '-file', "PushoverOut/BeamStress.out", '-closeOnWrite', '-

˓→ele', 102, 'section', '4', 'fiber','1', 'stressStrain')

328

329 # analyze(NstepsPush)
330

331 # Perform pushover analysis

332 dataPush = np.zeros((NstepsPush+1,5))
333 for j in range(NstepsPush):
334 analyze(1)
335 dataPush[j+1,0] = nodeDisp(31,1)
336 reactions()
337 dataPush[j+1,1] = nodeReaction(1, 1) + nodeReaction(2, 1) +
˓→nodeReaction(3, 1)

338

339 plt.plot(dataPush[:,0], -dataPush[:,1])

340 plt.xlim(0, MaxDisp)
341 plt.xticks(np.linspace(0,MaxDisp,5,endpoint=True))
342 plt.yticks(np.linspace(0, -int(dataPush[NstepsPush,1]),10,endpoint=True))
343 plt.grid(linestyle='dotted')
344 plt.xlabel('Top Displacement (inch)')
345 plt.ylabel('Base Shear (kip)')
346 plt.show()
347

348

349 print("Pushover analysis complete")

(continues on next page)

1.14. Examples 319

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

350

351

352

353

Cantilever FRP-Confined Circular Reinforced Concrete Column under Cyclic Lateral Loading

1. The source code is developed by Michael Haas & Konstantinos G. Megalooikonomou, German Research Centre
for Geosciences (GFZ)
2. The source code is shown below, which can be downloaded here.
3. Run the source code in your favorite Python program and should see following plot.

1.14.2 Earthquake Examples

1. Cantilever 2D EQ ground motion with gravity Analysis

2. Reinforced Concrete Frame Earthquake Analysis
3. Example name spaced nonlinear SDOF
4. RotD Spectra of Ground Motion
5. Portal 2D Frame - Dynamic EQ Ground Motion
6. 2D Column - Dynamic EQ Ground Motion
7. Nonlinear Canti Col Uniaxial Inelastic Section- Dyn EQ GM
8. Nonlin Canti Col Inelstc Uniaxial Mat in Fiber Sec - Dyn EQ
9. Cantilever 2D Column with Units- Dynamic EQ Ground Motion
10. Cantilever 2D Column with Units-Static Pushover
11. 2D Portal Frame with Units- Dynamic EQ Ground Motion
12. 2D Portal Frame with Units- Multiple Support Dynamic EQ Ground Motion-acctimeseries
13. 2D Portal Frame with Units- Multiple Support Dynamic EQ Ground Motion-disptimeseries
14. 2D Portal Frame with Units- Uniform Dynamic EQ -bidirectional-acctimeseries

Cantilever 2D EQ ground motion with gravity Analysis

1. The source code is shown below, which can be downloaded here.

2. The ground motion data file here must be put in the same folder.
3. Run the source code in your favorate Python program and should see results below

=========================================================
Start cantilever 2D EQ ground motion with gravity example
u2 = -0.07441860465116278
Passed!
=========================================

320 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1 print("=========================================================")
2 print("Start cantilever 2D EQ ground motion with gravity example")
3

4 from openseespy.opensees import *

5

7 # ------------------------------------------------------------------------------------
˓→--------------

8 # Example 1. cantilever 2D
9 # EQ ground motion with gravity
10 # all units are in kip, inch, second
11 # elasticBeamColumn ELEMENT
12 # Silvia Mazzoni & Frank McKenna, 2006
13 #
14 # ^Y
15 # |
16 # 2 __
17 # | |
18 # | |
19 # | |
20 # (1) 36'
21 # | |
22 # | |
23 # | |
24 # =1= ---- -------->X
25 #
26

27 # SET UP ----------------------------------------------------------------------------
28 wipe() # clear opensees model
29 model('basic', '-ndm', 2, '-ndf', 3) # 2 dimensions, 3 dof per node
30 # file mkdir data # create data directory
31

32 # define GEOMETRY -------------------------------------------------------------

33 # nodal coordinates:
34 node(1, 0., 0.) # node#, X Y
35 node(2, 0., 432.)
36

37 # Single point constraints -- Boundary Conditions

38 fix(1, 1, 1, 1) # node DX DY RZ
39

40 # nodal masses:
41 mass(2, 5.18, 0., 0.) # node#, Mx My Mz, Mass=Weight/g.
42

43 # Define ELEMENTS -------------------------------------------------------------

44 # define geometric transformation: performs a linear geometric transformation of beam
˓→stiffness and resisting force from the basic system to the global-coordinate system

45 geomTransf('Linear', 1) # associate a tag to transformation

46

47 # connectivity:
48 element('elasticBeamColumn', 1, 1, 2, 3600.0, 3225.0,1080000.0, 1)
49

50 # define GRAVITY -------------------------------------------------------------

51 timeSeries('Linear', 1)
52 pattern('Plain', 1, 1,)
53 load(2, 0., -2000., 0.) # node#, FX FY MZ --
˓→superstructure-weight

54

(continues on next page)

1.14. Examples 321

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

55 constraints('Plain') # how it handles boundary
˓→conditions

56 numberer('Plain') # renumber dof's to minimize band-width

˓→(optimization), if you want to

57 system('BandGeneral') # how to store and solve the system of

˓→equations in the analysis

58 algorithm('Linear') # use Linear algorithm for linear analysis

59 integrator('LoadControl', 0.1) # determine the next time step
˓→for an analysis, # apply gravity in 10 steps

60 analysis('Static') # define type of

˓→analysis static or transient

61 analyze(10) # perform gravity analysis

62 loadConst('-time', 0.0) # hold gravity constant and
˓→restart time

63

64 # DYNAMIC ground-motion analysis -----------------------------------------------------

˓→--------

65 # create load pattern

66 G = 386.0
67 timeSeries('Path', 2, '-dt', 0.005, '-filePath', 'A10000.dat', '-factor', G) # define
˓→acceleration vector from file (dt=0.005 is associated with the input file gm)

68 pattern('UniformExcitation', 2, 1, '-accel', 2) # define

˓→where and how (pattern tag, dof) acceleration is applied

69

70 # set damping based on first eigen mode

71 freq = eigen('-fullGenLapack', 1)[0]**0.5
72 dampRatio = 0.02
73 rayleigh(0., 0., 0., 2*dampRatio/freq)
74

75 # create the analysis

76 wipeAnalysis() # clear previously-define analysis
˓→parameters

77 constraints('Plain') # how it handles boundary conditions

78 numberer('Plain') # renumber dof's to minimize band-width (optimization), if you
˓→want to

79 system('BandGeneral') # how to store and solve the system of equations in the analysis
80 algorithm('Linear') # use Linear algorithm for linear analysis
81 integrator('Newmark', 0.5, 0.25) # determine the next time step for an analysis
82 analysis('Transient') # define type of analysis: time-dependent
83 analyze(3995, 0.01) # apply 3995 0.01-sec time steps in analysis
84

85 u2 = nodeDisp(2, 2)
86 print("u2 = ", u2)
87

88

89 if abs(u2+0.07441860465116277579) < 1e-12:

90 print("Passed!")
91 else:
92 print("Failed!")
93

94 wipe()
95

96 print("=========================================")

322 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Reinforced Concrete Frame Earthquake Analysis

1. The source code is shown below, which can be downloaded here.

2. The file for gravity analysis is also needed here.
3. The ReadRecord is a useful python function for parsing the PEER strong motion data base files and returning
the dt, nPts and creating a file containing just data points. The function is kept in a seperate file here and is
imported in the example.
4. The ground motion data file here must be put in the same folder.
5. Run the source code in your favorite Python program and should see Passed! in the results and a plotting of
displacement for node 3

1 print("==========================")
2 print("Start RCFrameEarthquake Example")
3

4 # Units: kips, in, sec

5 #
6 # Written: Minjie
7

8 from openseespy.opensees import *

9

10 import ReadRecord
11 import numpy as np
12 import matplotlib.pyplot as plt
(continues on next page)

1.14. Examples 323

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

13

14 wipe()
15 # ----------------------------------------------------
16 # Start of Model Generation & Initial Gravity Analysis
17 # ----------------------------------------------------
18

19 # Do operations of Example3.1 by sourcing in the tcl file

20 import RCFrameGravity
21 print("Gravity Analysis Completed")
22

23 # Set the gravity loads to be constant & reset the time in the domain
24 loadConst('-time', 0.0)
25

26 # ----------------------------------------------------
27 # End of Model Generation & Initial Gravity Analysis
28 # ----------------------------------------------------
29

30 # Define nodal mass in terms of axial load on columns

31 g = 386.4
32 m = RCFrameGravity.P/g
33

34 mass(3, m, m, 0.0)
35 mass(4, m, m, 0.0)
36

37 # Set some parameters

38 record = 'elCentro'
39

40 # Permform the conversion from SMD record to OpenSees record

41 dt, nPts = ReadRecord.ReadRecord(record+'.at2', record+'.dat')
42

43 # Set time series to be passed to uniform excitation

44 timeSeries('Path', 2, '-filePath', record+'.dat', '-dt', dt, '-factor', g)
45

46 # Create UniformExcitation load pattern

47 # tag dir
48 pattern('UniformExcitation', 2, 1, '-accel', 2)
49

50 # set the rayleigh damping factors for nodes & elements

51 rayleigh(0.0, 0.0, 0.0, 0.000625)
52

53 # Delete the old analysis and all it's component objects

54 wipeAnalysis()
55

56 # Create the system of equation, a banded general storage scheme

57 system('BandGeneral')
58

59 # Create the constraint handler, a plain handler as homogeneous boundary

60 constraints('Plain')
61

62 # Create the convergence test, the norm of the residual with a tolerance of
63 # 1e-12 and a max number of iterations of 10
64 test('NormDispIncr', 1.0e-12, 10 )
65

66 # Create the solution algorithm, a Newton-Raphson algorithm

67 algorithm('Newton')
68

69 # Create the DOF numberer, the reverse Cuthill-McKee algorithm

(continues on next page)

324 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

70 numberer('RCM')
71

72 # Create the integration scheme, the Newmark with alpha =0.5 and beta =.25
73 integrator('Newmark', 0.5, 0.25 )
74

75 # Create the analysis object

76 analysis('Transient')
77

78 # Perform an eigenvalue analysis

79 numEigen = 2
80 eigenValues = eigen(numEigen)
81 print("eigen values at start of transient:",eigenValues)
82

83 # set some variables

84 tFinal = nPts*dt
85 tCurrent = getTime()
86 ok = 0
87

88 time = [tCurrent]
89 u3 = [0.0]
90

91 # Perform the transient analysis

92 while ok == 0 and tCurrent < tFinal:
93

94 ok = analyze(1, .01)
95

96 # if the analysis fails try initial tangent iteration

97 if ok != 0:
98 print("regular newton failed .. lets try an initail stiffness for this step")
99 test('NormDispIncr', 1.0e-12, 100, 0)
100 algorithm('ModifiedNewton', '-initial')
101 ok =analyze( 1, .01)
102 if ok == 0:
103 print("that worked .. back to regular newton")
104 test('NormDispIncr', 1.0e-12, 10 )
105 algorithm('Newton')
106

107 tCurrent = getTime()

108

109 time.append(tCurrent)
110 u3.append(nodeDisp(3,1))
111

112

113

114 # Perform an eigenvalue analysis

115 eigenValues = eigen(numEigen)
116 print("eigen values at end of transient:",eigenValues)
117

118 results = open('results.out','a+')

119

120 if ok == 0:
121 results.write('PASSED : RCFrameEarthquake.py\n');
122 print("Passed!")
123 else:
124 results.write('FAILED : RCFrameEarthquake.py\n');
125 print("Failed!")
126
(continues on next page)

1.14. Examples 325

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

127 results.close()
128

129 plt.plot(time, u3)

130 plt.ylabel('Horizontal Displacement of node 3 (in)')
131 plt.xlabel('Time (s)')
132

133 plt.show()
134

135

136

137 print("==========================")

Example name spaced nonlinear SDOF

1. The source code is developed by Maxim Millen from University of Porto.

2. The source code is shown below, which can be downloaded here.
3. Also download the constants file here, and the ground motion file
4. Make sure the numpy, matplotlib and eqsig packages are installed in your Python distribution.
5. Run the source code in your favorite Python program and should see

326 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1 import eqsig
2 from eqsig import duhamels
3 import matplotlib.pyplot as plt
4 import numpy as np
5

6 import openseespy.opensees as op
7 import opensees_constants as opc #opensees_constants.py should be close to main file
˓→or use sys.path... to its directory
8

10 def get_inelastic_response(mass, k_spring, f_yield, motion, dt, xi=0.05, r_post=0.0):

11 """
12 Run seismic analysis of a nonlinear SDOF
13

14 :param mass: SDOF mass

15 :param k_spring: spring stiffness
16 :param f_yield: yield strength
17 :param motion: list, acceleration values
18 :param dt: float, time step of acceleration values
19 :param xi: damping ratio
20 :param r_post: post-yield stiffness
21 :return:
22 """
23

24 op.wipe()
25 op.model('basic', '-ndm', 2, '-ndf', 3) # 2 dimensions, 3 dof per node
26

27 # Establish nodes
28 bot_node = 1
29 top_node = 2
30 op.node(bot_node, 0., 0.)
31 op.node(top_node, 0., 0.)
32

33 # Fix bottom node

34 op.fix(top_node, opc.FREE, opc.FIXED, opc.FIXED)
35 op.fix(bot_node, opc.FIXED, opc.FIXED, opc.FIXED)
36 # Set out-of-plane DOFs to be slaved
37 op.equalDOF(1, 2, *[2, 3])
38

39 # nodal mass (weight / g):

40 op.mass(top_node, mass, 0., 0.)
41

42 # Define material
43 bilinear_mat_tag = 1
44 mat_type = "Steel01"
45 mat_props = [f_yield, k_spring, r_post]
46 op.uniaxialMaterial(mat_type, bilinear_mat_tag, *mat_props)
47

48 # Assign zero length element

49 beam_tag = 1
50 op.element('zeroLength', beam_tag, bot_node, top_node, "-mat", bilinear_mat_tag,
˓→"-dir", 1, '-doRayleigh', 1)

51

52 # Define the dynamic analysis

53 load_tag_dynamic = 1
54 pattern_tag_dynamic = 1
55

(continues on next page)

1.14. Examples 327

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

56 values = list(-1 * motion) # should be negative
57 op.timeSeries('Path', load_tag_dynamic, '-dt', dt, '-values', *values)
58 op.pattern('UniformExcitation', pattern_tag_dynamic, opc.X, '-accel', load_tag_
˓→dynamic)

59

60 # set damping based on first eigen mode

61 angular_freq = op.eigen('-fullGenLapack', 1) ** 0.5
62 alpha_m = 0.0
63 beta_k = 2 * xi / angular_freq
64 beta_k_comm = 0.0
65 beta_k_init = 0.0
66

67 op.rayleigh(alpha_m, beta_k, beta_k_init, beta_k_comm)

68

69 # Run the dynamic analysis

70

71 op.wipeAnalysis()
72

73 op.algorithm('Newton')
74 op.system('SparseGeneral')
75 op.numberer('RCM')
76 op.constraints('Transformation')
77 op.integrator('Newmark', 0.5, 0.25)
78 op.analysis('Transient')
79

80 tol = 1.0e-10
81 iterations = 10
82 op.test('EnergyIncr', tol, iterations, 0, 2)
83 analysis_time = (len(values) - 1) * dt
84 analysis_dt = 0.001
85 outputs = {
86 "time": [],
87 "rel_disp": [],
88 "rel_accel": [],
89 "rel_vel": [],
90 "force": []
91 }
92

93 while op.getTime() < analysis_time:

94 curr_time = op.getTime()
95 op.analyze(1, analysis_dt)
96 outputs["time"].append(curr_time)
97 outputs["rel_disp"].append(op.nodeDisp(top_node, 1))
98 outputs["rel_vel"].append(op.nodeVel(top_node, 1))
99 outputs["rel_accel"].append(op.nodeAccel(top_node, 1))
100 op.reactions()
101 outputs["force"].append(-op.nodeReaction(bot_node, 1)) # Negative since diff
˓→node

102 op.wipe()
103 for item in outputs:
104 outputs[item] = np.array(outputs[item])
105

106 return outputs

107

108

109 def show_single_comparison():

110 """
(continues on next page)

328 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

111 Create a plot of an elastic analysis, nonlinear analysis and closed form elastic
112

113 :return:
114 """
115

116 record_filename = 'test_motion_dt0p01.txt'

117 motion_step = 0.01
118 rec = np.loadtxt(record_filename)
119 acc_signal = eqsig.AccSignal(rec, motion_step)
120 period = 1.0
121 xi = 0.05
122 mass = 1.0
123 f_yield = 1.5 # Reduce this to make it nonlinear
124 r_post = 0.0
125

126 periods = np.array([period])

127 resp_u, resp_v, resp_a = duhamels.response_series(motion=rec, dt=motion_step,
˓→periods=periods, xi=xi)

128

129 k_spring = 4 * np.pi ** 2 * mass / period ** 2

130 outputs = get_inelastic_response(mass, k_spring, f_yield, rec, motion_step, xi=xi,
˓→ r_post=r_post)

131 outputs_elastic = get_inelastic_response(mass, k_spring, f_yield * 100, rec,

˓→motion_step, xi=xi, r_post=r_post)

132 ux_opensees = outputs["rel_disp"]

133 ux_opensees_elastic = outputs_elastic["rel_disp"]
134

135 bf, sps = plt.subplots(nrows=2)

136 sps[0].plot(acc_signal.time, resp_u[0], label="Eqsig")
137 sps[0].plot(outputs["time"], ux_opensees, label="Opensees fy=%.3gN" % f_yield, ls=
˓→"--")

138 sps[0].plot(outputs["time"], ux_opensees_elastic, label="Opensees fy=%.3gN" % (f_

˓→yield * 100), ls="--")

139 sps[1].plot(acc_signal.time, resp_a[0], label="Eqsig") # Elastic solution

140 time = acc_signal.time
141 acc_opensees_elastic = np.interp(time, outputs_elastic["time"], outputs_elastic[
˓→"rel_accel"]) - rec

142 print("diff", sum(acc_opensees_elastic - resp_a[0]))

143 sps[1].plot(time, acc_opensees_elastic, label="Opensees fy=%.2gN" % (f_yield *
˓→100), ls="--")

144 sps[0].legend()
145 sps[1].legend()
146 plt.show()
147

148

149 if __name__ == '__main__':

150 show_single_comparison()

RotD Spectra of Ground Motion

1. The source code is developed by Jawad Fayaz from University of California- Irvine.
2. The source code is shown below, which can be downloaded here.
3. Also download the code to read the provided GM file here.
4. The example bi-directional ground motion time histories are given GM11, GM21, GM12, GM22.

1.14. Examples 329

OpenSeesPy Documentation, Release 3.5.1.3

5. Run the source code in any Python IDE (e.g Spyder, Jupyter Notebook) and should see

330 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

"""
author : JAWAD FAYAZ (email: [email protected]) (website: https://jfayaz.github.io)

------------------------------ Instructions -------------------------------------

This code develops the RotD50 Sa and RotD100 Sa Spectra of the Bi-Directional
Ground Motion records as '.AT2' files provided in the current directory

The two directions of the ground motion record must be named as 'GM1i' and 'GM2i',
where 'i' is the ground motion number which goes from 1 to 'n', 'n' being the total
number of ground motions for which the Spectra needs to be generated. The extension
of the files must be '.AT2'

For example: If the Spectra of two ground motion records are required, 4 files with
the following names must be provided in the given 'GM' folder:
'GM11.AT2' - Ground Motion 1 in direction 1 (direction 1 can be either one of the
˓→bi-directional GM as we are rotating the ground motions it does not matter)

'GM21.AT2' - Ground Motion 1 in direction 2 (direction 2 is the other direction

˓→of the bi-directional GM)

'GM12.AT2' - Ground Motion 2 in direction 1 (direction 1 can be either one of the

˓→bi-directional GM as we are rotating the ground motions it does not matter)

'GM22.AT2' - Ground Motion 2 in direction 2 (direction 2 is the other direction

˓→of the bi-directional GM)

The Ground Motion file must be a vector file with 4 header lines.The first 3 lines
˓→can have

any content, however, the 4th header line must be written exactly as per the
˓→following example:

'NPTS= 15864, DT= 0.0050'

The 'ReadGMFile.py' can be edited accordingly for any other format

You may run this code in python IDE: 'Spyder' or any other similar IDE

Make sure you have the following python libraries installed:

os
sys
pathlib
fnmatch
shutil
IPython
pandas
numpy
matplotlib.pyplot

INPUT:
This codes provides the option to have 3 different regions of developing the Spectra
˓→of ground motions with different period intervals (discretizations)

The following inputs within the code are required:

'Path_to_openpyfiles'--> Path where the library files 'opensees.pyd' and 'LICENSE.
˓→rst' of OpenSeesPy are included (for further details go to https://openseespydoc.

˓→readthedocs.io/en/latest/windows.html)

'Int_T_Reg_1' --> Period Interval for the first region of the Spectrum
'End_T_Reg_1' --> Last Period of the first region of the Spectrum (where
˓→to end the first region)

'Int_T_Reg_2' --> Period Interval for the second region of the Spectrum
'End_T_Reg_2' --> Last Period of the second region of the Spectrum (where
˓→to end the second region)

'Int_T_Reg_3' --> Period Interval for the third region of the Spectrum
(continues on next page)

1.14. Examples 331

OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

'End_T_Reg_3' --> Last Period of the third region of the Spectrum (where
˓→to end the third region)

'Plot_Spectra' --> whether to plot the generated Spectra of the ground

˓→motions (options: 'Yes', 'No')

OUTPUT:
The output will be provided in a saperate 'GMi_Spectra.txt' file for each ground
˓→motion record, where 'i' denotes the number of ground motion in the same of

provided 'GM1i.AT2' and 'GM2i.AT2' files. The output files will be generated in a
˓→saperate folder 'Spectra' which will be created in the current folder

The 'GMi_Spectra.txt' file will consist of space-saperated file with:

'Periods (secs)' 'RotD50 Sa (g)' 'RotD100 Sa (g)'

%%%%%
˓→===================================================================================================

˓→%%%%%%

%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
˓→%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

˓→%%%%%%%%%%%%

"""

##### ================== INPUTS ================== #####

# Path where the library files 'opensees.pyd' and 'LICENSE.rst' are included (for
˓→further details go to https://openseespydoc.readthedocs.io/en/latest/windows.html)

Path_to_openpyfiles = 'C:\Tcl'

# For periods 0 to 'End_T_Reg_1' in an interval of 'Int_T_Reg_1'

Int_T_Reg_1 = 0.1
End_T_Reg_1 = 1

# For periods ['End_T_Reg_1'+'Int_T_Reg_2'] to 'End_T_Reg_2' in an interval of 'Int_T_

˓→Reg_2'

Int_T_Reg_2 = 0.2
End_T_Reg_2 = 2

# For periods ['End_T_Reg_2'+'Int_T_Reg_3'] to 'End_T_Reg_3' in an interval of 'Int_T_

˓→Reg_3'

Int_T_Reg_3 = 0.5
End_T_Reg_3 = 5

# Plot Spectra (options: 'Yes' or 'No')

Plot_Spectra = 'Yes'

##### =============== CODE BEGINS ================ #######

## Importing Libraries
import os, sys, pathlib, fnmatch
import shutil as st
from IPython import get_ipython

from openseespy.opensees import *

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
(continues on next page)

332 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

import warnings
import matplotlib.cbook
warnings.filterwarnings("ignore",category=matplotlib.cbook.mplDeprecation)
wipe()

# Getting Number of Ground Motions from the GM folder

GMdir = os.getcwd()
No_of_GMs = int(len(fnmatch.filter(os.listdir(GMdir),'*.AT2'))/2)
print('\nGenerating Spectra for {} provided GMs \n\n'.format(np.round(No_of_GMs,0)))

# Initializations
DISPLACEMENTS = pd.DataFrame(columns=['uX','uY'])
GM_SPECTRA = pd.DataFrame(columns=['Period(s)','RotD50Sa(g)', 'RotD100Sa(g)'])
SDOF_RESPONSE = [[]]
GM_RESPONSE = [[]]

# Spectra Generation
for iEQ in range(1,No_of_GMs+1):
print('Generating Spectra for GM: {} ...\n'.format(np.round(iEQ,0)))
Periods = np.concatenate((list(np.arange(Int_T_Reg_1,End_T_Reg_1+Int_T_Reg_1,Int_
˓→T_Reg_1)),list(np.arange(End_T_Reg_1+Int_T_Reg_2,End_T_Reg_2+Int_T_Reg_2,Int_T_Reg_

˓→2)),list(np.arange(End_T_Reg_2+Int_T_Reg_3,End_T_Reg_3+Int_T_Reg_3,Int_T_Reg_3))),

˓→axis=0)

ii = 0

for T in Periods:
ii = ii+1
GMinter = 0

# Storing Periods
GM_SPECTRA.loc[ii-1,'Period(s)'] = T

# Setting modelbuilder
model('basic', '-ndm', 3, '-ndf', 6)

# Setting SODF Variables

g = 386.1 # value of g
L = 1.0 # Length
d = 2 # Diameter
r = d/2 # Radius
A = np.pi*(r**2) # Area
E = 1.0 # Elastic Modulus
G = 1.0 # Shear Modulus
I3 = np.pi*(r**4)/4 # Moment of Inertia (zz)
J = np.pi*(r**4)/2 # Polar Moment of Inertia
I2 = np.pi*(r**4)/4 # Moment of Inertia (yy)
K = 3*E*I3/(L**3) # Stiffness
M = K*(T**2)/4/(np.pi**2) # Mass
omega = np.sqrt(K/M) # Natural Frequency
Tn = 2*np.pi/omega # Natural Period

# Creating nodes
node(1, 0.0, 0.0, 0.0)
node(2, 0.0, 0.0, L)

# Transformation
transfTag = 1
(continues on next page)

1.14. Examples 333

OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

geomTransf('Linear',transfTag,0.0,1.0,0.0)

# Setting boundary condition

fix(1, 1, 1, 1, 1, 1, 1)

# Defining materials
uniaxialMaterial("Elastic", 11, E)

# Defining elements
element("elasticBeamColumn",12,1,2,A,E,G,J,I2,I3,1)

# Defining mass
mass(2,M,M,0.0,0.0,0.0,0.0)

# Eigen Value Analysis (Verifying Period)

numEigen = 1
eigenValues = eigen(numEigen)
omega = np.sqrt(eigenValues)
T = 2*np.pi/omega
print(' Calculating Spectral Ordinate for Period = {} secs'.format(np.
˓→round(T,3)))

## Reading GM Files
exec(open("ReadGMFile.py").read()) # read in procedure
˓→Multinition

iGMinput = 'GM1'+str(iEQ)+' GM2'+str(iEQ) ;

GMinput = iGMinput.split(' ');
gmXY = {}
for i in range(0,2):
inFile = GMdir + '\\'+ GMinput[i]+'.AT2';
dt, NumPts , gmXY = ReadGMFile()

# Storing GM Histories
gmX = gmXY[1]
gmY = gmXY[2]
gmXY_mat = np.column_stack((gmX,gmX,gmY,gmY))

# Bidirectional Uniform Earthquake ground motion (uniform acceleration input

˓→at all support nodes)
iGMfile = 'GM1'+str(iEQ)+' GM2'+str(iEQ) ;
GMfile = iGMfile.split(' ')
GMdirection = [1,1,2,2];
GMfact = [np.cos(GMinter*np.pi/180),np.sin(-GMinter*np.pi/180),
˓→np.sin(GMinter*np.pi/180), np.cos(GMinter*np.pi/180)];

IDTag = 2
loop = [1,2,3,4]

for i in loop:
# Setting time series to be passed to uniform excitation
timeSeries('Path',IDTag +i, '-dt', dt, '-values', *list(gmXY_mat[:,i-1]),
˓→'-factor', GMfact[i-1]*g)

# Creating UniformExcitation load pattern

pattern('UniformExcitation', IDTag+i, GMdirection[i-1], '-accel',
˓→IDTag+i)

# Defining Damping
# Applying Rayleigh Damping from $xDamp
(continues on next page)

334 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

# D=$alphaM*M + $betaKcurr*Kcurrent + $betaKcomm*KlastCommit + $beatKinit*
˓→$Kinitial

xDamp = 0.05;
˓→ # 5% damping ratio
alphaM = 0.;
˓→ # M-prop. damping;
˓→D = alphaM*M

betaKcurr = 0.;
˓→ # K-proportional damping;
˓→+beatKcurr*KCurrent

betaKcomm = 2.*xDamp/omega; # K-
˓→prop. damping parameter; +betaKcomm*KlastCommitt
betaKinit = 0.;
˓→ # initial-stiffness proportional
˓→damping +beatKinit*Kini
rayleigh(alphaM,betaKcurr,betaKinit,betaKcomm); # RAYLEIGH damping

# Creating the analysis

wipeAnalysis() # clear previously-define
˓→analysis parameters

constraints("Penalty",1e18, 1e18) # how to handle boundary conditions

numberer("RCM") # renumber dof's to minimize band-width
˓→(optimization), if you want to

system('SparseGeneral') # how to store and solve the system of

˓→equations in the analysis

algorithm('Linear') # use Linear algorithm for linear

˓→analysis

integrator("TRBDF2") # determine the next time step for an

˓→analysis

algorithm("NewtonLineSearch") # define type of analysis: time-dependent

test('EnergyIncr',1.0e-6, 100, 0)
analysis("Transient")

# Variables (Can alter the speed of analysis)

dtAnalysis = dt
TmaxAnanlysis = dt*NumPts
tFinal = int(TmaxAnanlysis/dtAnalysis)
tCurrent = getTime()
ok = 0
time = [tCurrent]

# Initializations of response
u1 = [0.0]
u2 = [0.0]

# Performing the transient analysis (Performance is slow in this loop, can be

˓→altered by changing the parameters)
while ok == 0 and tCurrent < tFinal:
ok = analyze(1, dtAnalysis)
# if the analysis fails try initial tangent iteration
if ok != 0:
print("Iteration failed .. lets try an initial stiffness for this step
˓→")

test('NormDispIncr', 1.0e-12, 100, 0)

algorithm('ModifiedNewton', '-initial')
ok =analyze( 1, .001)

(continues on next page)

1.14. Examples 335

OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

if ok == 0:
print("that worked .. back to regular newton")
test('NormDispIncr', 1.0e-12, 10 )
algorithm('Newton')

tCurrent = getTime()
time.append(tCurrent)
u1.append(nodeDisp(2,1))
u2.append(nodeDisp(2,2))

# Storing responses
DISPLACEMENTS.loc[ii-1,'uX'] = np.array(u1)
DISPLACEMENTS.loc[ii-1,'uY'] = np.array(u2)
DISP_X_Y = np.column_stack((np.array(u1),np.array(u2)))

# Rotating the Spectra (Projections)

Rot_Matrix = np.zeros((2,2))
Rot_Disp = np.zeros((180,1))
for theta in range (0,180,1):
Rot_Matrix [0,0] = np.cos(np.deg2rad(theta))
Rot_Matrix [0,1] = np.sin(np.deg2rad(-theta))
Rot_Matrix [1,0] = np.sin(np.deg2rad(theta))
Rot_Matrix [1,1] = np.cos(np.deg2rad(theta))
Rot_Disp[theta,0] = np.max(np.matmul(DISP_X_Y,Rot_Matrix)[:,0])

# Storing Spectra
Rot_Acc = np.dot(Rot_Disp,(omega**2)/g)
GM_SPECTRA.loc[ii-1,'RotD50Sa(g)'] = np.median(Rot_Acc)
GM_SPECTRA.loc[ii-1,'RotD100Sa(g)']= np.max(Rot_Acc)
wipe()

# Writing Spectra to Files

if not os.path.exists('Spectra'):
os.makedirs('Spectra')
GM_SPECTRA.to_csv('Spectra//GM'+str(iEQ)+'_Spectra.txt', sep=' ',header=True,
˓→index=False)

# Plotting Spectra
if Plot_Spectra == 'Yes':

def plot_spectra(PlotTitle,SpectraType,iGM):
axes = fig.add_subplot(1, 1, 1)
axes.plot(GM_SPECTRA['Period(s)'] , GM_SPECTRA[SpectraType] , '.-',lw=7,
˓→markersize=20, label='GM'+str(iGM))

axes.set_xlabel('Period (sec)',fontsize=30,fontweight='bold')
axes.set_ylabel(SpectraType,fontsize=30,fontweight='bold')
axes.set_title(PlotTitle,fontsize=40,fontweight='bold')
axes.tick_params(labelsize= 25)
axes.grid(True)
axes.set_xlim(0, np.ceil(max(GM_SPECTRA['Period(s)'])))
axes.set_ylim(0, np.ceil(max(GM_SPECTRA[SpectraType])))
axes.axhline(linewidth=10,color='black')
axes.axvline(linewidth=10,color='black')
axes.hold(True)
axes.legend(fontsize =30)

fig = plt.figure(1,figsize=(18,12))
(continues on next page)

336 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

plot_spectra('RotD50 Spectra','RotD50Sa(g)',iEQ)

fig = plt.figure(2,figsize=(18,12))
plot_spectra('RotD100 Spectra','RotD100Sa(g)',iEQ)

SDOF_RESPONSE.insert(iEQ-1,DISPLACEMENTS)
GM_RESPONSE.insert(iEQ-1,GM_SPECTRA)

print('\nGenerated Spectra for GM: {}\n\n'.format(np.round(iEQ,0)))

Portal 2D Frame - Dynamic EQ Ground Motion

Converted to openseespy by: Pavan Chigullapally

University of Auckland
Email: [email protected]

1. This is a simple model of an elastic portal frame with EQ ground motion and gravity loading. Here the structure
is excited using uniform excitation load pattern
2. All units are in kip, inch, second
3. To run EQ ground-motion analysis, BM68elc.acc needs to be downloaded into the same directory)
4. The source code is shown below, which can be downloaded here.
5. The detailed problem description can be found here (example: 1b)

1 # -*- coding: utf-8 -*-

2 """
3 Created on Mon Apr 22 17:29:26 2019
4

5 @author: pchi893
6 """
7 # Converted to openseespy by: Pavan Chigullapally
8 # University of Auckland
9 # Email: [email protected]
10 # Example 1b. portal frame in 2D
11 #This is a simple model of an elastic portal frame with EQ ground motion and gravity
˓→loading. Here the structure is excited using uniform excitation load pattern

12 # all units are in kip, inch, second

13 #To run EQ ground-motion analysis (BM68elc.acc needs to be downloaded into the same
˓→directory).

14 #the detailed problem description can be found here: http://opensees.berkeley.edu/

˓→wiki/index.php/Examples_Manual (example: 1b)
15 # ------------------------------------------------------------------------------------
˓→--------------

16 # elasticBeamColumn ELEMENT
17 # OpenSees (Tcl) code by: Silvia Mazzoni & Frank McKenna, 2006
18

19 #
20 # ^Y
21 # |
22 # 3_________(3)________4 __
23 # | | |
24 # | | |
25 # | | |
(continues on next page)

1.14. Examples 337

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

26 # (1) (2) LCol
27 # | | |
28 # | | |
29 # | | |
30 # =1= =2= _|_ -------->X
31 # |----------LBeam------------|
32 #
33

34 # SET UP -----------------------------------------------------------------------------
35

36 import openseespy.opensees as op
37 #import the os module
38 import os
39 op.wipe()
40

41 ######################################################################################
˓→###################################################################################

42

43 ######################################################################################
˓→###################################################################################

44 op.model('basic', '-ndm', 2, '-ndf', 3)

45

46 #to create a directory at specified path with name "Data"

47 #os.chdir('C:\\Opensees Python\\OpenseesPy examples')
48

49 #this will create the directory with name 'Data' and will update it when we rerun the
˓→analysis, otherwise we have to keep deleting the old 'Data' Folder

50 dir = "Data-1b"
51 if not os.path.exists(dir):
52 os.makedirs(dir)
53

54 #this will create just 'Data' folder

55 #os.mkdir("Data-1b")
56

57 #detect the current working directory

58 #path1 = os.getcwd()
59 #print(path1)
60

61 h = 432.0
62 w = 504.0
63

64 op.node(1, 0.0, 0.0)

65 op.node(2, h, 0.0)
66 op.node(3, 0.0, w)
67 op.node(4, h, w)
68

69 op.fix(1, 1,1,1)
70 op.fix(2, 1,1,1)
71 op.fix(3, 0,0,0)
72 op.fix(4, 0,0,0)
73

74 op.mass(3, 5.18, 0.0, 0.0)

75 op.mass(4, 5.18, 0.0, 0.0)
76

77 op.geomTransf('Linear', 1)
78 A = 3600000000.0
79 E = 4227.0
(continues on next page)

338 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

80 Iz = 1080000.0
81

82 A1 = 5760000000.0
83 Iz1 = 4423680.0
84 op.element('elasticBeamColumn', 1, 1, 3, A, E, Iz, 1)
85 op.element('elasticBeamColumn', 2, 2, 4, A, E, Iz, 1)
86 op.element('elasticBeamColumn', 3, 3, 4, A1, E, Iz1, 1)
87

88 op.recorder('Node', '-file', 'Data-1b/DFree.out','-time', '-node', 3,4, '-dof', 1,2,3,

˓→ 'disp')

89 op.recorder('Node', '-file', 'Data-1b/DBase.out','-time', '-node', 1,2, '-dof', 1,2,3,

˓→ 'disp')

90 op.recorder('Node', '-file', 'Data-1b/RBase.out','-time', '-node', 1,2, '-dof', 1,2,3,

˓→ 'reaction')

91 #op.recorder('Drift', '-file', 'Data-1b/Drift.out','-time', '-node', 1, '-dof', 1,2,3,

˓→ 'disp')

92 op.recorder('Element', '-file', 'Data-1b/FCol.out','-time', '-ele', 1,2, 'globalForce

˓→')

93 op.recorder('Element', '-file', 'Data-1b/DCol.out','-time', '-ele', 3, 'deformations')

94

95 #defining gravity loads

96 op.timeSeries('Linear', 1)
97 op.pattern('Plain', 1, 1)
98 op.eleLoad('-ele', 3, '-type', '-beamUniform', -7.94)
99

100 op.constraints('Plain')
101 op.numberer('Plain')
102 op.system('BandGeneral')
103 op.test('NormDispIncr', 1e-8, 6)
104 op.algorithm('Newton')
105 op.integrator('LoadControl', 0.1)
106 op.analysis('Static')
107 op.analyze(10)
108

109 op.loadConst('-time', 0.0)

110

111 #applying Dynamic Ground motion analysis

112 op.timeSeries('Path', 2, '-dt', 0.01, '-filePath', 'BM68elc.acc', '-factor', 1.0)
113 op.pattern('UniformExcitation', 2, 1, '-accel', 2) #how to give accelseriesTag?
114

115 eigen = op. eigen('-fullGenLapack', 1)

116 import math
117 power = math.pow(eigen[0], 0.5)
118 betaKcomm = 2 * (0.02/power)
119

120 op.rayleigh(0.0, 0.0, 0.0, betaKcomm)

121

122 op.wipeAnalysis()
123 op.constraints('Plain')
124 op.numberer('Plain')
125 op.system('BandGeneral')
126 op.test('NormDispIncr', 1e-8, 10)
127 op.algorithm('Newton')
128 op.integrator('Newmark', 0.5, 0.25)
129 op.analysis('Transient')
130 op.analyze(1000, 0.02)
131
(continues on next page)

1.14. Examples 339

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

132 u3 = op.nodeDisp(3, 1)
133 print("u2 = ", u3)
134

135 op.wipe()

2D Column - Dynamic EQ Ground Motion

Converted to openseespy by: Pavan Chigullapally

University of Auckland
Email: [email protected]

1. EQ ground motion with gravity- uniform excitation of structure

2. All units are in kip, inch, second
3. Note: In this example, all input values for Example 1a are replaced by variables. The objective of this example
is to demonstrate the use of variables in defining
4. The OpenSees input and also to run various tests and algorithms at once to increase the chances of convergence
5. To run EQ ground-motion analysis (BM68elc.acc needs to be downloaded into the same directory)
6. The detailed problem description can be found here (example:2a)
7. The source code is shown below, which can be downloaded here.
1 # -*- coding: utf-8 -*-
2 """
3 Created on Mon Apr 22 15:12:06 2019
4

5 @author: pchi893
6 """
7 # Converted to openseespy by: Pavan Chigullapally
8 # University of Auckland
9 # Email: [email protected]
10

11 # EQ ground motion with gravity- uniform excitation of structure

12 # all units are in kip, inch, second
13 ##Note: In this example, all input values for Example 1a are replaced by variables.
˓→The objective of this example is to demonstrate the use of variables in defining

14 #the OpenSees input and also to run various tests and algorithms at once to increase
˓→the chances of convergence

15 # Example 2a. 2D cantilever column, dynamic eq ground motion

16 #To run EQ ground-motion analysis (BM68elc.acc needs to be downloaded into the same
˓→directory)

17 #the detailed problem description can be found here: http://opensees.berkeley.edu/

˓→wiki/index.php/Examples_Manual (example:2a)
18 # ------------------------------------------------------------------------------------
˓→--------------

19 # OpenSees (Tcl) code by: Silvia Mazzoni & Frank McKenna, 2006
20

21 #
22 # ^Y
23 # |
24 # 2 __
25 # | |
26 # | |
(continues on next page)

340 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

27 # | |
28 # (1) LCol
29 # | |
30 # | |
31 # | |
32 # =1= _|_ -------->X
33 #
34

35 # SET UP ----------------------------------------------------------------------------
36 import openseespy.opensees as op
37 #import the os module
38 import os
39 import math
40 op.wipe()
41

42 ######################################################################################
˓→###################################################################################

43

44 ######################################################################################
˓→###################################################################################

45 op.model('basic', '-ndm', 2, '-ndf', 3)

46

47 #to create a directory at specified path with name "Data"

48

49

50 #this will create the directory with name 'Data' and will update it when we rerun the
˓→analysis, otherwise we have to keep deleting the old 'Data' Folder

51 dir = "Data-2a"
52 if not os.path.exists(dir):
53 os.makedirs(dir)
54

55 #this will create just 'Data' folder

56 #os.mkdir("Data")
57

58 #detect the current working directory

59 #path1 = os.getcwd()
60 #print(path1)
61

62 LCol = 432.0 # column length

63 Weight = 2000.0 # superstructure weight
64

65 # define section geometry

66 HCol = 60.0 # Column Depth
67 BCol = 60.0 # Column Width
68

69 PCol =Weight # nodal dead-load weight per column

70 g = 386.4
71 Mass = PCol/g
72

73 ACol = HCol*BCol*1000 # cross-sectional area, make stiff

74 IzCol = (BCol*math.pow(HCol,3))/12 # Column moment of inertia
75

76 op.node(1, 0.0, 0.0)

77 op.node(2, 0.0, LCol)
78

79 op.fix(1, 1, 1, 1)
80
(continues on next page)

1.14. Examples 341

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

81 op.mass(2, Mass, 1e-9, 0.0)
82

83 ColTransfTag = 1
84 op.geomTransf('Linear', ColTransfTag)
85 #A = 3600000000.0
86 #E = 4227.0
87 #Iz = 1080000.0
88

89 fc = -4.0 # CONCRETE Compressive Strength (+Tension, -Compression)

90 Ec = 57*math.sqrt(-fc*1000) # Concrete Elastic Modulus (the term in sqr root needs to
˓→be in psi

91

92 op.element('elasticBeamColumn', 1, 1, 2, ACol, Ec, IzCol, ColTransfTag)

93

94 op.recorder('Node', '-file', 'Data-2a/DFree.out','-time', '-node', 2, '-dof', 1,2,3,

˓→'disp')

95 op.recorder('Node', '-file', 'Data-2a/DBase.out','-time', '-node', 1, '-dof', 1,2,3,

˓→'disp')

96 op.recorder('Node', '-file', 'Data-2a/RBase.out','-time', '-node', 1, '-dof', 1,2,3,

˓→'reaction')

97 #op.recorder('Drift', '-file', 'Data-2a/Drift.out','-time', '-node', 1, '-dof', 1,2,3,

˓→ 'disp')

98 op.recorder('Element', '-file', 'Data-2a/FCol.out','-time', '-ele', 1, 'globalForce')

99 op.recorder('Element', '-file', 'Data-2a/DCol.out','-time', '-ele', 1, 'deformations')
100

101 #defining gravity loads

102 op.timeSeries('Linear', 1)
103 op.pattern('Plain', 1, 1)
104 op.load(2, 0.0, -PCol, 0.0)
105

106 Tol = 1e-8 # convergence tolerance for test

107 NstepGravity = 10
108 DGravity = 1/NstepGravity
109 op.integrator('LoadControl', DGravity) # determine the next time step for an analysis
110 op.numberer('Plain') # renumber dof's to minimize band-width (optimization), if you
˓→want to

111 op.system('BandGeneral') # how to store and solve the system of equations in the
˓→analysis

112 op.constraints('Plain') # how it handles boundary conditions

113 op.test('NormDispIncr', Tol, 6) # determine if convergence has been achieved at the
˓→end of an iteration step

114 op.algorithm('Newton') # use Newton's solution algorithm: updates tangent stiffness

˓→at every iteration

115 op.analysis('Static') # define type of analysis static or transient

116 op.analyze(NstepGravity) # apply gravity
117

118 op.loadConst('-time', 0.0) #maintain constant gravity loads and reset time to zero
119

120 #applying Dynamic Ground motion analysis

121 GMdirection = 1
122 GMfile = 'BM68elc.acc'
123 GMfact = 1.0
124

125

126

127 Lambda = op.eigen('-fullGenLapack', 1) # eigenvalue mode 1

128 import math
(continues on next page)

342 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

129 Omega = math.pow(Lambda[0], 0.5)
130 betaKcomm = 2 * (0.02/Omega)
131

132 xDamp = 0.02 # 2% damping ratio

133 alphaM = 0.0 # M-prop. damping; D = alphaM*M
134 betaKcurr = 0.0 # K-proportional damping; +beatKcurr*KCurrent
135 betaKinit = 0.0 # initial-stiffness proportional damping +beatKinit*Kini
136

137 op.rayleigh(alphaM,betaKcurr, betaKinit, betaKcomm) # RAYLEIGH damping

138

139 # Uniform EXCITATION: acceleration input

140 IDloadTag = 400 # load tag
141 dt = 0.01 # time step for input ground motion
142 GMfatt = 1.0 # data in input file is in g Unifts --
˓→ACCELERATION TH

143 maxNumIter = 10
144 op.timeSeries('Path', 2, '-dt', dt, '-filePath', GMfile, '-factor', GMfact)
145 op.pattern('UniformExcitation', IDloadTag, GMdirection, '-accel', 2)
146

147 op.wipeAnalysis()
148 op.constraints('Transformation')
149 op.numberer('Plain')
150 op.system('BandGeneral')
151 op.test('EnergyIncr', Tol, maxNumIter)
152 op.algorithm('ModifiedNewton')
153

154 NewmarkGamma = 0.5

155 NewmarkBeta = 0.25
156 op.integrator('Newmark', NewmarkGamma, NewmarkBeta)
157 op.analysis('Transient')
158

159 DtAnalysis = 0.01

160 TmaxAnalysis = 10.0
161

162 Nsteps = int(TmaxAnalysis/ DtAnalysis)

163

164 ok = op.analyze(Nsteps, DtAnalysis)

165 tCurrent = op.getTime()
166

167 # for gravity analysis, load control is fine, 0.1 is the load factor increment (http:/
˓→/opensees.berkeley.edu/wiki/index.php/Load_Control)

168

169 test = {1:'NormDispIncr', 2: 'RelativeEnergyIncr', 4: 'RelativeNormUnbalance',5:

˓→'RelativeNormDispIncr', 6: 'NormUnbalance'}

170 algorithm = {1:'KrylovNewton', 2: 'SecantNewton' , 4: 'RaphsonNewton',5:

˓→'PeriodicNewton', 6: 'BFGS', 7: 'Broyden', 8: 'NewtonLineSearch'}

171

172 for i in test:

173 for j in algorithm:
174

175 if ok != 0:
176 if j < 4:
177 op.algorithm(algorithm[j], '-initial')
178

179 else:
180 op.algorithm(algorithm[j])
181
(continues on next page)

1.14. Examples 343

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

182 op.test(test[i], Tol, 1000)
183 ok = op.analyze(Nsteps, DtAnalysis)
184 print(test[i], algorithm[j], ok)
185 if ok == 0:
186 break
187 else:
188 continue
189

190 u2 = op.nodeDisp(2, 1)
191 print("u2 = ", u2)
192

193 op.wipe()

Nonlinear Canti Col Uniaxial Inelastic Section- Dyn EQ GM

Converted to openseespy by: Pavan Chigullapally

University of Auckland
Email: [email protected]

1. EQ ground motion with gravity- uniform excitation of structure

2. The nonlinear beam-column element that replaces the elastic element of Example 2a requires the definition of
the element cross section, or its behavior. In this example,
3. The Uniaxial Section used to define the nonlinear moment-curvature behavior of the element section is “aggre-
gated” to an elastic response for the axial behavior to define
4. The required characteristics of the column element in the 2D model. In a 3D model, torsional behavior would
also have to be aggregated to this section.
5. Note:In this example, both the axial behavior (typically elastic) and the flexural behavior (moment curvature)
are defined indepenently and are then “aggregated” into a section.
6. This is a characteristic of the uniaxial section: there is no coupling of behaviors.
7. To run EQ ground-motion analysis (BM68elc.acc needs to be downloaded into the same directory)
8. The problem description can be found here (example:2b)
9. The source code is shown below, which can be downloaded here.
1 # -*- coding: utf-8 -*-
2 """
3 Created on Mon Apr 22 15:12:06 2019
4

5 @author: pchi893
6 """
7 # Converted to openseespy by: Pavan Chigullapally
8 # University of Auckland
9 # Email: [email protected]
10 # Example 2b. 2D cantilever column, dynamic eq ground motion
11 # EQ ground motion with gravity- uniform excitation of structure
12 #he nonlinear beam-column element that replaces the elastic element of Example 2a
˓→requires the definition of the element cross section, or its behavior. In this

˓→example,

13 #the Uniaxial Section used to define the nonlinear moment-curvature behavior of the
˓→element section is "aggregated" to an elastic response for the axial behavior to

˓→define
(continues on next page)

344 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

14 #the required characteristics of the column element in the 2D model. In a 3D model,
˓→torsional behavior would also have to be aggregated to this section.

15 #Note:In this example, both the axial behavior (typically elastic) and the flexural
˓→behavior (moment curvature) are defined indepenently and are then "aggregated" into

˓→a section.

16 #This is a characteristic of the uniaxial section: there is no coupling of behaviors.

17

18 #To run EQ ground-motion analysis (BM68elc.acc needs to be downloaded into the same
˓→directory)

19 #the problem description can be found here: http://opensees.berkeley.edu/wiki/index.

˓→php/Examples_Manual(example:2b)

20 # ------------------------------------------------------------------------------------
˓→--------------

21 # OpenSees (Tcl) code by: Silvia Mazzoni & Frank McKenna, 2006
22 #
23 # ^Y
24 # |
25 # 2 __
26 # | |
27 # | |
28 # | |
29 # (1) LCol
30 # | |
31 # | |
32 # | |
33 # =1= _|_ -------->X
34 #
35

36 # SET UP ----------------------------------------------------------------------------
37 import openseespy.opensees as op
38 #import the os module
39 import os
40 import math
41 op.wipe()
42 ######################################################################################
˓→###################################################################################

43

44 #to create a directory at specified path with name "Data"

45 os.chdir('C:\\Opensees Python\\OpenseesPy examples')
46

47 #this will create the directory with name 'Data' and will update it when we rerun the
˓→analysis, otherwise we have to keep deleting the old 'Data' Folder

48 dir = "C:\\Opensees Python\\OpenseesPy examples\\Data-2b"

49 if not os.path.exists(dir):
50 os.makedirs(dir)
51 #this will create just 'Data' folder
52 #os.mkdir("Data")
53 #detect the current working directory
54 #path1 = os.getcwd()
55 #print(path1)
56 ######################################################################################
˓→###################################################################################

57

58 ######################################################################################
˓→###################################################################################

59 op.model('basic', '-ndm', 2, '-ndf', 3)

60 LCol = 432.0 # column length
(continues on next page)

1.14. Examples 345

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

61 Weight = 2000.0 # superstructure weight
62

63 # define section geometry

64 HCol = 60.0 # Column Depth
65 BCol = 60.0 # Column Width
66

67 PCol =Weight # nodal dead-load weight per column

68 g = 386.4
69 Mass = PCol/g
70

71 ACol = HCol*BCol*1000 # cross-sectional area, make stiff

72 IzCol = (BCol*math.pow(HCol,3))/12 # Column moment of inertia
73

74 op.node(1, 0.0, 0.0)

75 op.node(2, 0.0, LCol)
76

77 op.fix(1, 1, 1, 1)
78

79 op.mass(2, Mass, 1e-9, 0.0)

80

81 #Define Elements and Sections

82 ColMatTagFlex = 2
83 ColMatTagAxial = 3
84 ColSecTag = 1
85 BeamSecTag = 2
86

87 fc = -4.0 # CONCRETE Compressive Strength (+Tension, -Compression)

88 Ec = 57*math.sqrt(-fc*1000) # Concrete Elastic Modulus (the term in sqr root needs to
˓→be in psi

89

90 #Column Section
91 EICol = Ec*IzCol # EI, for moment-curvature relationship
92 EACol = Ec*ACol # EA, for axial-force-strain relationship
93 MyCol = 130000.0 #yield Moment calculated
94 PhiYCol = 0.65e-4 # yield curvature
95 EIColCrack = MyCol/PhiYCol # cracked section inertia
96 b = 0.01 # strain-hardening ratio (ratio between post-yield tangent and initial
˓→elastic tangent)

97

98 op.uniaxialMaterial('Steel01', ColMatTagFlex, MyCol, EIColCrack, b) #steel moment

˓→curvature isused for Mz of the section only, # bilinear behavior for flexure

99 op.uniaxialMaterial('Elastic', ColMatTagAxial, EACol) # this is not used as a

˓→material, this is an axial-force-strain response

100 op.section('Aggregator', ColSecTag, ColMatTagAxial, 'P', ColMatTagFlex, 'Mz') #

˓→combine axial and flexural behavior into one section (no P-M interaction here)

101

102 ColTransfTag = 1
103 op.geomTransf('Linear', ColTransfTag)
104 numIntgrPts = 5
105 eleTag = 1
106 op.element('nonlinearBeamColumn', eleTag, 1, 2, numIntgrPts, ColSecTag, ColTransfTag)
107

108 op.recorder('Node', '-file', 'Data-2b/DFree.out','-time', '-node', 2, '-dof', 1,2,3,

˓→'disp')

109 op.recorder('Node', '-file', 'Data-2b/DBase.out','-time', '-node', 1, '-dof', 1,2,3,

˓→'disp')

110 op.recorder('Node', '-file', 'Data-2b/RBase.out','-time', '-node', 1, '-dof', 1,2,3,

˓→'reaction') (continues on next page)

346 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

111 #op.recorder('Drift', '-file', 'Data-2b/Drift.out','-time', '-node', 1, '-dof', 1,2,3,
˓→ 'disp')

112 op.recorder('Element', '-file', 'Data-2b/FCol.out','-time', '-ele', 1, 'globalForce')

113 op.recorder('Element', '-file', 'Data-2b/ForceColSec1.out','-time', '-ele', 1,
˓→'section', 1, 'force')

114 #op.recorder('Element', '-file', 'Data-2b/DCol.out','-time', '-ele', 1, 'deformations

˓→')

115

116 #defining gravity loads

117 op.timeSeries('Linear', 1)
118 op.pattern('Plain', 1, 1)
119 op.load(2, 0.0, -PCol, 0.0)
120

121 Tol = 1e-8 # convergence tolerance for test

122 NstepGravity = 10
123 DGravity = 1/NstepGravity
124 op.integrator('LoadControl', DGravity) # determine the next time step for an analysis
125 op.numberer('Plain') # renumber dof's to minimize band-width (optimization), if you
˓→want to

126 op.system('BandGeneral') # how to store and solve the system of equations in the
˓→analysis

127 op.constraints('Plain') # how it handles boundary conditions

128 op.test('NormDispIncr', Tol, 6) # determine if convergence has been achieved at the
˓→end of an iteration step

129 op.algorithm('Newton') # use Newton's solution algorithm: updates tangent stiffness

˓→at every iteration

130 op.analysis('Static') # define type of analysis static or transient

131 op.analyze(NstepGravity) # apply gravity
132

133 op.loadConst('-time', 0.0) #maintain constant gravity loads and reset time to zero
134

135 #applying Dynamic Ground motion analysis

136 GMdirection = 1
137 GMfile = 'BM68elc.acc'
138 GMfact = 1.0
139

140

141

142 Lambda = op.eigen('-fullGenLapack', 1) # eigenvalue mode 1

143 import math
144 Omega = math.pow(Lambda, 0.5)
145 betaKcomm = 2 * (0.02/Omega)
146

147 xDamp = 0.02 # 2% damping ratio

148 alphaM = 0.0 # M-prop. damping; D = alphaM*M
149 betaKcurr = 0.0 # K-proportional damping; +beatKcurr*KCurrent
150 betaKinit = 0.0 # initial-stiffness proportional damping +beatKinit*Kini
151

152 op.rayleigh(alphaM,betaKcurr, betaKinit, betaKcomm) # RAYLEIGH damping

153

154 # Uniform EXCITATION: acceleration input

155 IDloadTag = 400 # load tag
156 dt = 0.01 # time step for input ground motion
157 GMfatt = 1.0 # data in input file is in g Unifts --
˓→ACCELERATION TH

158 maxNumIter = 10
159 op.timeSeries('Path', 2, '-dt', dt, '-filePath', GMfile, '-factor', GMfact)
(continues on next page)

1.14. Examples 347

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

160 op.pattern('UniformExcitation', IDloadTag, GMdirection, '-accel', 2)
161

162 op.wipeAnalysis()
163 op.constraints('Transformation')
164 op.numberer('Plain')
165 op.system('BandGeneral')
166 op.test('EnergyIncr', Tol, maxNumIter)
167 op.algorithm('ModifiedNewton')
168

169 NewmarkGamma = 0.5

170 NewmarkBeta = 0.25
171 op.integrator('Newmark', NewmarkGamma, NewmarkBeta)
172 op.analysis('Transient')
173

174 DtAnalysis = 0.01

175 TmaxAnalysis = 10.0
176

177 Nsteps = int(TmaxAnalysis/ DtAnalysis)

178

179 ok = op.analyze(Nsteps, DtAnalysis)

180

181 tCurrent = op.getTime()

182

183 # for gravity analysis, load control is fine, 0.1 is the load factor increment (http:/
˓→/opensees.berkeley.edu/wiki/index.php/Load_Control)

184

185 test = {1:'NormDispIncr', 2: 'RelativeEnergyIncr', 4: 'RelativeNormUnbalance',5:

˓→'RelativeNormDispIncr', 6: 'NormUnbalance'}

186 algorithm = {1:'KrylovNewton', 2: 'SecantNewton' , 4: 'RaphsonNewton',5:

˓→'PeriodicNewton', 6: 'BFGS', 7: 'Broyden', 8: 'NewtonLineSearch'}

187

188 for i in test:

189 for j in algorithm:
190

191 if ok != 0:
192 if j < 4:
193 op.algorithm(algorithm[j], '-initial')
194

195 else:
196 op.algorithm(algorithm[j])
197

198 op.test(test[i], Tol, 1000)

199 ok = op.analyze(Nsteps, DtAnalysis)
200 print(test[i], algorithm[j], ok)
201 if ok == 0:
202 break
203 else:
204 continue
205

206 u2 = op.nodeDisp(2, 1)
207 print("u2 = ", u2)
208

209 op.wipe()

348 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

Nonlin Canti Col Inelstc Uniaxial Mat in Fiber Sec - Dyn EQ

Converted to openseespy by: Pavan Chigullapally

University of Auckland
Email: [email protected]

1. EQ ground motion with gravity- uniform excitation of structure

2. In this example, the Uniaxial Section of Example 2b is replaced by a fiber section. Inelastic uniaxial materials
are used in this example,
3. Which are assigned to each fiber, or patch of fibers, in the section.
4. In this example the axial and flexural behavior are coupled, a characteristic of the fiber section.
5. The nonlinear/inelastic behavior of a fiber section is defined by the stress-strain response of the uniaxial materials
used to define it.
6. To run EQ ground-motion analysis (BM68elc.acc needs to be downloaded into the same directory)
7. The problem description can be found here (example:2c)
8. The source code is shown below, which can be downloaded here.
1 # -*- coding: utf-8 -*-
2 """
3 Created on Mon Apr 22 15:12:06 2019
4

5 @author: pchi893
6 """
7 # Converted to openseespy by: Pavan Chigullapally
8 # University of Auckland
9 # Email: [email protected]
10 # Example 2c. 2D cantilever column, dynamic eq ground motion
11 # EQ ground motion with gravity- uniform excitation of structure
12 #In this example, the Uniaxial Section of Example 2b is replaced by a fiber section.
˓→Inelastic uniaxial materials are used in this example,

13 #which are assigned to each fiber, or patch of fibers, in the section.

14 #In this example the axial and flexural behavior are coupled, a characteristic of the
˓→fiber section.

15 #The nonlinear/inelastic behavior of a fiber section is defined by the stress-strain

˓→response of the uniaxial materials used to define it.

16

17 #To run EQ ground-motion analysis (BM68elc.acc needs to be downloaded into the same
˓→directory)

18 #the problem description can be found here: http://opensees.berkeley.edu/wiki/index.

˓→php/Examples_Manual(example: 2c)

19 # ------------------------------------------------------------------------------------
˓→--------------

20 # OpenSees (Tcl) code by: Silvia Mazzoni & Frank McKenna, 2006
21

22 #
23 # ^Y
24 # |
25 # 2 __
26 # | |
27 # | |
28 # | |
29 # (1) LCol
30 # | |
(continues on next page)

1.14. Examples 349

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

31 # | |
32 # | |
33 # =1= _|_ -------->X
34 #
35

36 # SET UP ----------------------------------------------------------------------------
37 import openseespy.opensees as op
38 #import the os module
39 import os
40 import math
41 op.wipe()
42 ######################################################################################
˓→###################################################################################

43 #to create a directory at specified path with name "Data"

44 os.chdir('C:\\Opensees Python\\OpenseesPy examples')
45

46 #this will create the directory with name 'Data' and will update it when we rerun the
˓→analysis, otherwise we have to keep deleting the old 'Data' Folder

47 dir = "C:\\Opensees Python\\OpenseesPy examples\\Data-2c"

48 if not os.path.exists(dir):
49 os.makedirs(dir)
50 #this will create just 'Data' folder
51 #os.mkdir("Data")
52 #detect the current working directory
53 #path1 = os.getcwd()
54 #print(path1)
55 ######################################################################################
˓→###################################################################################

56

57 ######################################################################################
˓→###################################################################################

58 op.model('basic', '-ndm', 2, '-ndf', 3)

59 LCol = 432.0 # column length
60 Weight = 2000.0 # superstructure weight
61

62 # define section geometry

63 HCol = 60.0 # Column Depth
64 BCol = 60.0 # Column Width
65

66 PCol =Weight # nodal dead-load weight per column

67 g = 386.4
68 Mass = PCol/g
69

70 ACol = HCol*BCol*1000 # cross-sectional area, make stiff

71 IzCol = (BCol*math.pow(HCol,3))/12 # Column moment of inertia
72

73 op.node(1, 0.0, 0.0)

74 op.node(2, 0.0, LCol)
75

76 op.fix(1, 1, 1, 1)
77

78 op.mass(2, Mass, 1e-9, 0.0)

79

80 ColSecTag = 1 # assign a tag number to the column section

81 coverCol = 5.0 # Column cover to reinforcing steel NA.
82 numBarsCol = 16 # number of longitudinal-reinforcement bars in column. (symmetric
˓→top & bot)
(continues on next page)

350 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

83 barAreaCol = 2.25 # area of longitudinal-reinforcement bars
84

85 # MATERIAL parameters
86 IDconcU = 1 # material ID tag -- unconfined cover concrete
˓→(here used for complete section)

87 IDreinf = 2 # material ID tag -- reinforcement

88

89 # nominal concrete compressive strength

90 fc = -4.0 # CONCRETE Compressive Strength (+Tension, -
˓→Compression)

91 Ec = 57*math.sqrt(-fc*1000) # Concrete Elastic Modulus (the term in sqr root needs to

˓→be in psi

92

93 # unconfined concrete
94 fc1U = fc # UNCONFINED concrete (todeschini parabolic model),
˓→maximum stress

95 eps1U = -0.003 # strain at maximum strength of unconfined

˓→concrete

96 fc2U = 0.2*fc1U # ultimate stress

97 eps2U = -0.01 # strain at ultimate stress
98 Lambda = 0.1 # ratio between unloading slope at $eps2
˓→and initial slope $Ec

99

100 # tensile-strength properties

101 ftU = -0.14* fc1U # tensile strength +tension
102 Ets = ftU/0.002 # tension softening stiffness
103

104 Fy = 66.8 # STEEL yield stress

105 Es = 29000.0 # modulus of steel
106 Bs = 0.01 # strain-hardening ratio
107 R0 = 18.0 # control the transition from elastic to
˓→plastic branches

108 cR1 = 0.925 # control the transition from elastic to

˓→plastic branches

109 cR2 = 0.15 # control the transition from elastic to

˓→plastic branches

110

111 op.uniaxialMaterial('Concrete02', IDconcU, fc1U, eps1U, fc2U, eps2U, Lambda, ftU,

˓→Ets) # build cover concrete (unconfined)

112 op.uniaxialMaterial('Steel02', IDreinf, Fy, Es, Bs, R0,cR1,cR2) # build reinforcement

˓→material

113 # FIBER SECTION properties -----------------------------------------------------------

˓→--

114 # symmetric section

115 # y
116 # ^
117 # |
118 # --------------------- -- --
119 # | o o o | | -- cover
120 # | | |
121 # | | |
122 # z <--- | + | H
123 # | | |
124 # | | |
125 # | o o o | | -- cover
126 # --------------------- -- --
127 # |-------- B --------|
(continues on next page)

1.14. Examples 351

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

128 #
129 # RC section:
130

131 coverY = HCol/2.0 # The distance from the section z-axis to the edge of the
˓→cover concrete -- outer edge of cover concrete

132 coverZ = BCol/2.0 # The distance from the section y-axis to the edge of the
˓→cover concrete -- outer edge of cover concrete

133 coreY = coverY-coverCol

134 coreZ = coverZ-coverCol
135 nfY = 16 # number of fibers for concrete in y-direction
136 nfZ = 4 # number of fibers for concrete in z-direction
137

138 op.section('Fiber', ColSecTag)

139 op.patch('quad', IDconcU, nfZ, nfY, -coverY,coverZ, -coverY,-coverZ, coverY,-coverZ,
˓→coverY,coverZ) # Define the concrete patch

140 op.layer('straight', IDreinf, numBarsCol, barAreaCol, -coreY,coreZ,-coreY,-coreZ)

141 op.layer('straight', IDreinf, numBarsCol, barAreaCol, coreY,coreZ, coreY,-coreZ)
142 ColTransfTag = 1
143 op.geomTransf('Linear', ColTransfTag)
144 numIntgrPts = 5
145 eleTag = 1
146

147 #import InelasticFiberSection

148

149 op.element('nonlinearBeamColumn', eleTag, 1, 2, numIntgrPts, ColSecTag, ColTransfTag)

150

151 op.recorder('Node', '-file', 'Data-2c/DFree.out','-time', '-node', 2, '-dof', 1,2,3,

˓→'disp')

152 op.recorder('Node', '-file', 'Data-2c/DBase.out','-time', '-node', 1, '-dof', 1,2,3,

˓→'disp')

153 op.recorder('Node', '-file', 'Data-2c/RBase.out','-time', '-node', 1, '-dof', 1,2,3,

˓→'reaction')

154 #op.recorder('Drift', '-file', 'Data-2c/Drift.out','-time', '-node', 1, '-dof', 1,2,3,

˓→ 'disp')

155 op.recorder('Element', '-file', 'Data-2c/FCol.out','-time', '-ele', 1, 'globalForce')

156 op.recorder('Element', '-file', 'Data-2c/ForceColSec1.out','-time', '-ele', 1,
˓→'section', 1, 'force')

157 #op.recorder('Element', '-file', 'Data-2c/DCol.out','-time', '-ele', 1, 'deformations

˓→')

158

159 #defining gravity loads

160 op.timeSeries('Linear', 1)
161 op.pattern('Plain', 1, 1)
162 op.load(2, 0.0, -PCol, 0.0)
163

164 Tol = 1e-8 # convergence tolerance for test

165 NstepGravity = 10
166 DGravity = 1/NstepGravity
167 op.integrator('LoadControl', DGravity) # determine the next time step for an analysis
168 op.numberer('Plain') # renumber dof's to minimize band-width (optimization), if you
˓→want to

169 op.system('BandGeneral') # how to store and solve the system of equations in the
˓→analysis

170 op.constraints('Plain') # how it handles boundary conditions

171 op.test('NormDispIncr', Tol, 6) # determine if convergence has been achieved at the
˓→end of an iteration step

172 op.algorithm('Newton') # use Newton's solution algorithm: updates tangent stiffness

˓→at every iteration (continues on next page)

352 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

173 op.analysis('Static') # define type of analysis static or transient
174 op.analyze(NstepGravity) # apply gravity
175

176 op.loadConst('-time', 0.0) #maintain constant gravity loads and reset time to zero
177

178 #applying Dynamic Ground motion analysis

179 GMdirection = 1
180 GMfile = 'BM68elc.acc'
181 GMfact = 1.0
182

183

184

185 Lambda = op.eigen('-fullGenLapack', 1) # eigenvalue mode 1

186 import math
187 Omega = math.pow(Lambda, 0.5)
188 betaKcomm = 2 * (0.02/Omega)
189

190 xDamp = 0.02 # 2% damping ratio

191 alphaM = 0.0 # M-prop. damping; D = alphaM*M
192 betaKcurr = 0.0 # K-proportional damping; +beatKcurr*KCurrent
193 betaKinit = 0.0 # initial-stiffness proportional damping +beatKinit*Kini
194

195 op.rayleigh(alphaM,betaKcurr, betaKinit, betaKcomm) # RAYLEIGH damping

196

197 # Uniform EXCITATION: acceleration input

198 IDloadTag = 400 # load tag
199 dt = 0.01 # time step for input ground motion
200 GMfatt = 1.0 # data in input file is in g Unifts --
˓→ACCELERATION TH

201 maxNumIter = 10
202 op.timeSeries('Path', 2, '-dt', dt, '-filePath', GMfile, '-factor', GMfact)
203 op.pattern('UniformExcitation', IDloadTag, GMdirection, '-accel', 2)
204

205 op.wipeAnalysis()
206 op.constraints('Transformation')
207 op.numberer('Plain')
208 op.system('BandGeneral')
209 op.test('EnergyIncr', Tol, maxNumIter)
210 op.algorithm('ModifiedNewton')
211

212 NewmarkGamma = 0.5

213 NewmarkBeta = 0.25
214 op.integrator('Newmark', NewmarkGamma, NewmarkBeta)
215 op.analysis('Transient')
216

217 DtAnalysis = 0.01

218 TmaxAnalysis = 10.0
219

220 Nsteps = int(TmaxAnalysis/ DtAnalysis)

221

222 ok = op.analyze(Nsteps, DtAnalysis)

223

224 tCurrent = op.getTime()

225

226 # for gravity analysis, load control is fine, 0.1 is the load factor increment (http:/
˓→/opensees.berkeley.edu/wiki/index.php/Load_Control)

227
(continues on next page)

1.14. Examples 353

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

228 test = {1:'NormDispIncr', 2: 'RelativeEnergyIncr', 4: 'RelativeNormUnbalance',5:
˓→'RelativeNormDispIncr', 6: 'NormUnbalance'}

229 algorithm = {1:'KrylovNewton', 2: 'SecantNewton' , 4: 'RaphsonNewton',5:

˓→'PeriodicNewton', 6: 'BFGS', 7: 'Broyden', 8: 'NewtonLineSearch'}

230

231 for i in test:

232 for j in algorithm:
233

234 if ok != 0:
235 if j < 4:
236 op.algorithm(algorithm[j], '-initial')
237

238 else:
239 op.algorithm(algorithm[j])
240

241 op.test(test[i], Tol, 1000)

242 ok = op.analyze(Nsteps, DtAnalysis)
243 print(test[i], algorithm[j], ok)
244 if ok == 0:
245 break
246 else:
247 continue
248

249 u2 = op.nodeDisp(2, 1)
250 print("u2 = ", u2)
251

252 op.wipe()

Cantilever 2D Column with Units- Dynamic EQ Ground Motion

Converted to openseespy by: Pavan Chigullapally

University of Auckland
Email: [email protected]

1. To run Uniaxial Inelastic Material, Fiber Section, Nonlinear Mode, Uniform Earthquake Excitation
2. First import the InelasticFiberSection.py (upto gravity loading is already in this script) and run the
current script
3. To run EQ ground-motion analysis BM68elc.acc needs to be downloaded into the same directory)
4. Same acceleration input at all nodes restrained in specified direction (uniform acceleration input at all support
nodes)
5. The problem description can be found here (example:3)
6. The source code is shown below, which can be downloaded here.

1 # -*- coding: utf-8 -*-

2 """
3 Created on Mon Apr 22 15:12:06 2019
4

5 @author: pchi893
6 """
7 # Converted to openseespy by: Pavan Chigullapally
8 # University of Auckland
(continues on next page)

354 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

9 # Email: [email protected]
10 # Example 3. 2D Cantilever -- EQ ground motion
11 #To run Uniaxial Inelastic Material, Fiber Section, Nonlinear Mode, Uniform
˓→Earthquake Excitation:First import the InelasticFiberSection.py(upto gravity

˓→loading is already in this script)

12 #and run the current script

13 #To run EQ ground-motion analysis (BM68elc.acc needs to be downloaded into the same
˓→directory)

14 # Same acceleration input at all nodes restrained in specified direction (uniform

˓→acceleration input at all support nodes)

15 #the detailed problem description can be found here: http://opensees.berkeley.edu/

˓→wiki/index.php/Examples_Manual (example: 3)
16 # ------------------------------------------------------------------------------------
˓→--------------

17 # OpenSees (Tcl) code by: Silvia Mazzoni & Frank McKenna, 2006
18 ######################################################################################
˓→####################################################################################

19 import openseespy.opensees as op
20 #import the os module
21 #import os
22 import math
23 op.wipe()
24 ######################################################################################
˓→###################################################################################

25 import InelasticFiberSection
26 #applying Dynamic Ground motion analysis
27 Tol = 1e-8
28 GMdirection = 1
29 GMfile = 'BM68elc.acc'
30 GMfact = 1.0
31 Lambda = op.eigen('-fullGenLapack', 1) # eigenvalue mode 1
32 Omega = math.pow(Lambda, 0.5)
33 betaKcomm = 2 * (0.02/Omega)
34

35 xDamp = 0.02 # 2% damping ratio

36 alphaM = 0.0 # M-prop. damping; D = alphaM*M
37 betaKcurr = 0.0 # K-proportional damping; +beatKcurr*KCurrent
38 betaKinit = 0.0 # initial-stiffness proportional damping +beatKinit*Kini
39

40 op.rayleigh(alphaM,betaKcurr, betaKinit, betaKcomm) # RAYLEIGH damping

41

42 # Uniform EXCITATION: acceleration input

43 IDloadTag = 400 # load tag
44 dt = 0.01 # time step for input ground motion
45 GMfatt = 1.0 # data in input file is in g Unifts --
˓→ACCELERATION TH

46 maxNumIter = 10
47 op.timeSeries('Path', 2, '-dt', dt, '-filePath', GMfile, '-factor', GMfact)
48 op.pattern('UniformExcitation', IDloadTag, GMdirection, '-accel', 2)
49

50 op.wipeAnalysis()
51 op.constraints('Transformation')
52 op.numberer('Plain')
53 op.system('BandGeneral')
54 op.test('EnergyIncr', Tol, maxNumIter)
55 op.algorithm('ModifiedNewton')
56
(continues on next page)

1.14. Examples 355

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

57 NewmarkGamma = 0.5
58 NewmarkBeta = 0.25
59 op.integrator('Newmark', NewmarkGamma, NewmarkBeta)
60 op.analysis('Transient')
61

62 DtAnalysis = 0.01 # time-step Dt for lateral analysis

63 TmaxAnalysis = 10.0 # maximum duration of ground-motion analysis
64

65 Nsteps = int(TmaxAnalysis/ DtAnalysis)

66

67 ok = op.analyze(Nsteps, DtAnalysis)
68

69 tCurrent = op.getTime()
70

71 # for gravity analysis, load control is fine, 0.1 is the load factor increment (http:/
˓→/opensees.berkeley.edu/wiki/index.php/Load_Control)

72

73 test = {1:'NormDispIncr', 2: 'RelativeEnergyIncr', 4: 'RelativeNormUnbalance',5:

˓→'RelativeNormDispIncr', 6: 'NormUnbalance'}

74 algorithm = {1:'KrylovNewton', 2: 'SecantNewton' , 4: 'RaphsonNewton',5:

˓→'PeriodicNewton', 6: 'BFGS', 7: 'Broyden', 8: 'NewtonLineSearch'}

75

76 for i in test:
77 for j in algorithm:
78

79 if ok != 0:
80 if j < 4:
81 op.algorithm(algorithm[j], '-initial')
82

83 else:
84 op.algorithm(algorithm[j])
85

86 op.test(test[i], Tol, 1000)

87 ok = op.analyze(Nsteps, DtAnalysis)
88 print(test[i], algorithm[j], ok)
89 if ok == 0:
90 break
91 else:
92 continue
93

94 u2 = op.nodeDisp(2, 1)
95 print("u2 = ", u2)
96

97 op.wipe()

Cantilever 2D Column with Units-Static Pushover

Converted to openseespy by: Pavan Chigullapally

University of Auckland
Email: [email protected]

1. To run Uniaxial Inelastic Material, Fiber Section, Nonlinear Mode, Static Pushover Analysis
2. First import the InelasticFiberSection.py (upto gravity loading is already in this script) and run the
current script

356 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

3. To run EQ ground-motion analysis BM68elc.acc needs to be downloaded into the same directory)
4. Same acceleration input at all nodes restrained in specified direction (uniform acceleration input at all support
nodes)
5. The problem description can be found here (example:3)
6. The source code is shown below, which can be downloaded here.

1 # -*- coding: utf-8 -*-

2 """
3 Created on Mon Apr 22 15:12:06 2019
4

5 @author: pchi893
6 """
7 # Converted to openseespy by: Pavan Chigullapally
8 # University of Auckland
9 # Email: [email protected]
10 # Example 3. 2D Cantilever -- Static Pushover
11 #To run Uniaxial Inelastic Material, Fiber Section, Nonlinear Mode, Static Pushover
˓→Analysis: First import the InelasticFiberSection.py(upto gravity loading is already

˓→in this script)

12 #and run the current script

13 #the detailed problem description can be found here: http://opensees.berkeley.edu/
˓→wiki/index.php/Examples_Manual (example: 3)
14 # ------------------------------------------------------------------------------------
˓→--------------

15 # OpenSees (Tcl) code by: Silvia Mazzoni & Frank McKenna, 2006
16 # characteristics of pushover analysis
17 ######################################################################################
˓→####################################################################################

18 import openseespy.opensees as op
19 #import the os module
20 #import os
21 import math
22 op.wipe()
23

24 from InelasticFiberSection import *

25 Dmax = 0.05*LCol
26 Dincr = 0.001*LCol
27 Hload = Weight
28 maxNumIter = 6
29 tol = 1e-8
30

31 op.timeSeries('Linear', 2)
32 op.pattern('Plain', 200, 2)
33 op.load(2, Hload, 0.0,0.0)
34

35 op.wipeAnalysis()
36 op.constraints('Plain')
37 op.numberer('Plain')
38 op.system('BandGeneral')
39 op.test('EnergyIncr', Tol, maxNumIter)
40 op.algorithm('Newton')
41

42 op.integrator('DisplacementControl', IDctrlNode, IDctrlDOF, Dincr)

43 op.analysis('Static')
44

45

(continues on next page)

1.14. Examples 357

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

46 Nsteps = int(Dmax/ Dincr)
47

48 ok = op.analyze(Nsteps)
49 print(ok)
50

51 # for gravity analysis, load control is fine, 0.1 is the load factor increment (http:/
˓→/opensees.berkeley.edu/wiki/index.php/Load_Control)

52

53 test = {1:'NormDispIncr', 2: 'RelativeEnergyIncr', 4: 'RelativeNormUnbalance',5:

˓→'RelativeNormDispIncr', 6: 'NormUnbalance'}

54 algorithm = {1:'KrylovNewton', 2: 'SecantNewton' , 4: 'RaphsonNewton',5:

˓→'PeriodicNewton', 6: 'BFGS', 7: 'Broyden', 8: 'NewtonLineSearch'}

55

56 for i in test:
57 for j in algorithm:
58

59 if ok != 0:
60 if j < 4:
61 op.algorithm(algorithm[j], '-initial')
62

63 else:
64 op.algorithm(algorithm[j])
65

66 op.test(test[i], Tol, 1000)

67 ok = op.analyze(Nsteps)
68 print(test[i], algorithm[j], ok)
69 if ok == 0:
70 break
71 else:
72 continue
73

74 u2 = op.nodeDisp(2, 1)
75 print("u2 = ", u2)
76

77 op.wipe()

2D Portal Frame with Units- Dynamic EQ Ground Motion

Converted to openseespy by: Pavan Chigullapally

University of Auckland
Email: [email protected]

1. To run Uniaxial Inelastic Material, Fiber Section, Nonlinear Mode, Uniform Earthquake Excitation
2. First import the InelasticFiberSectionPortal2Dframe.py
3. To run EQ ground-motion analysis (ReadRecord.py, H-E12140.AT2 needs to be downloaded into the
same directory)
4. Same acceleration input at all nodes restrained in specified direction (uniform acceleration input at all support
nodes)
5. The problem description can be found here (example:4)
6. The source code is shown below, which can be downloaded here.

358 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1 # -*- coding: utf-8 -*-

2 """
3 Created on Mon Apr 22 15:12:06 2019
4

5 @author: pchi893
6 """
7 # Converted to openseespy by: Pavan Chigullapally
8 # University of Auckland
9 # Email: [email protected]
10 # Example4. 2D Portal Frame-- Dynamic EQ input analysis
11

12 #To run Uniaxial Inelastic Material, Fiber Section, Nonlinear Mode, Uniform
˓→Earthquake Excitation:First import the InelasticFiberSectionPortal2Dframe.py

13 #(upto gravity loading is already in this script) and run the current script
14 #To run EQ ground-motion analysis (ReadRecord.py, H-E12140.AT2 needs to be downloaded
˓→into the same directory)

15 #Same acceleration input at all nodes restrained in specified direction (uniform

˓→acceleration input at all support nodes)

16 #the problem description can be found here:

17 #http://opensees.berkeley.edu/wiki/index.php/Examples_Manual and http://opensees.
˓→berkeley.edu/wiki/index.php/OpenSees_Example_4._Portal_Frame(example: 4)

18 # ------------------------------------------------------------------------------------
˓→--------------

19 # OpenSees (Tcl) code by: Silvia Mazzoni & Frank McKenna, 2006
20 ######################################################################################
˓→####################################################################################

21 import openseespy.opensees as op
22 #import the os module
23 #import os
24 import math
25 op.wipe()
26 ######################################################################################
˓→####################################################################################

27

28 from InelasticFiberSectionPortal2Dframe import *

29 #applying Dynamic Ground motion analysis
30 Tol = 1e-8
31 maxNumIter = 10
32 GMdirection = 1
33 GMfact = 1.5
34 GMfatt = g*GMfact
35 DtAnalysis = 0.01*sec # time-step Dt for lateral analysis
36 TmaxAnalysis = 10.0*sec # maximum duration of ground-motion analysis
37

38

39 Lambda = op.eigen('-fullGenLapack', 1) # eigenvalue mode 1

40 Omega = math.pow(Lambda, 0.5)
41 betaKcomm = 2 * (0.02/Omega)
42

43 xDamp = 0.02 # 2% damping ratio

44 alphaM = 0.0 # M-prop. damping; D = alphaM*M
45 betaKcurr = 0.0 # K-proportional damping; +beatKcurr*KCurrent
46 betaKinit = 0.0 # initial-stiffness proportional damping +beatKinit*Kini
47

48 op.rayleigh(alphaM,betaKcurr, betaKinit, betaKcomm) # RAYLEIGH damping

49

50 # Set some parameters

(continues on next page)

1.14. Examples 359

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

51 record = 'H-E12140'
52

53 import ReadRecord
54 # Permform the conversion from SMD record to OpenSees record
55 dt, nPts = ReadRecord.ReadRecord(record+'.at2', record+'.dat')
56 #print(dt, nPts)
57

58 # Uniform EXCITATION: acceleration input

59 IDloadTag = 400 # load tag
60 op.timeSeries('Path', 2, '-dt', dt, '-filePath', record+'.dat', '-factor', GMfatt)
61 op.pattern('UniformExcitation', IDloadTag, GMdirection, '-accel', 2)
62

63 op.wipeAnalysis()
64 op.constraints('Transformation')
65 op.numberer('RCM')
66 op.system('BandGeneral')
67 #op.test('EnergyIncr', Tol, maxNumIter)
68 #op.algorithm('ModifiedNewton')
69 #NewmarkGamma = 0.5
70 #NewmarkBeta = 0.25
71 #op.integrator('Newmark', NewmarkGamma, NewmarkBeta)
72 #op.analysis('Transient')
73

74

75 #Nsteps = int(TmaxAnalysis/ DtAnalysis)

76 #
77 #ok = op.analyze(Nsteps, DtAnalysis)
78

79 tCurrent = op.getTime()
80

81 # for gravity analysis, load control is fine, 0.1 is the load factor increment (http:/
˓→/opensees.berkeley.edu/wiki/index.php/Load_Control)

82

83 test = {1:'NormDispIncr', 2: 'RelativeEnergyIncr', 3:'EnergyIncr', 4:

˓→'RelativeNormUnbalance',5: 'RelativeNormDispIncr', 6: 'NormUnbalance'}

84 algorithm = {1:'KrylovNewton', 2: 'SecantNewton' , 3:'ModifiedNewton' , 4:

˓→'RaphsonNewton',5: 'PeriodicNewton', 6: 'BFGS', 7: 'Broyden', 8: 'NewtonLineSearch'}

85

86 tFinal = nPts*dt
87

88 #tFinal = 10.0*sec
89 time = [tCurrent]
90 u3 = [0.0]
91 u4 = [0.0]
92 ok = 0
93 while tCurrent < tFinal:
94 # ok = op.analyze(1, .01)
95 for i in test:
96 for j in algorithm:
97 if j < 4:
98 op.algorithm(algorithm[j], '-initial')
99

100 else:
101 op.algorithm(algorithm[j])
102 while ok == 0 and tCurrent < tFinal:
103

104 op.test(test[i], Tol, maxNumIter)

(continues on next page)

360 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

105 NewmarkGamma = 0.5
106 NewmarkBeta = 0.25
107 op.integrator('Newmark', NewmarkGamma, NewmarkBeta)
108 op.analysis('Transient')
109 ok = op.analyze(1, .01)
110

111 if ok == 0 :
112 tCurrent = op.getTime()
113 time.append(tCurrent)
114 u3.append(op.nodeDisp(3,1))
115 u4.append(op.nodeDisp(4,1))
116 print(test[i], algorithm[j], 'tCurrent=', tCurrent)
117

118

119 import matplotlib.pyplot as plt

120 plt.figure(figsize=(8,8))
121 plt.plot(time, u3)
122 plt.ylabel('Horizontal Displacement of node 3 (in)')
123 plt.xlabel('Time (s)')
124 plt.savefig('Horizontal Disp at Node 3 vs time.jpeg', dpi = 500)
125 plt.show()
126

127 plt.figure(figsize=(8,8))
128 plt.plot(time, u4)
129 plt.ylabel('Horizontal Displacement of node 4 (in)')
130 plt.xlabel('Time (s)')
131 plt.savefig('Horizontal Disp at Node 4 vs time.jpeg', dpi = 500)
132 plt.show()
133

134

135 op.wipe()

2D Portal Frame with Units- Multiple Support Dynamic EQ Ground Motion-acctimeseries

Converted to openseespy by: Pavan Chigullapally

University of Auckland
Email: [email protected]

1. To run Uniaxial Inelastic Material, Fiber Section, Nonlinear Mode, MultipleSupport Earthquake ground motion
2. First import the InelasticFiberSectionPortal2Dframe.py
3. Upto gravity loading is already in this script and run the current script
4. To run EQ ground-motion analysis (ReadRecord.py, H-E12140.AT2 needs to be downloaded into the
same directory)
5. MultipleSupport Earthquake ground motion (different acceleration input at specified support nodes) – two nodes
here
6. The problem description can be found here (example:4)
7. The source code is shown below, which can be downloaded here.
1 # -*- coding: utf-8 -*-
2 """
3 Created on Mon Apr 22 15:12:06 2019
(continues on next page)

1.14. Examples 361

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

4

5 @author: pchi893
6 """
7 # Converted to openseespy by: Pavan Chigullapally
8 # University of Auckland
9 # Email: [email protected]
10 # Example4. 2D Portal Frame-- Dynamic EQ input analysis-- multiple-support
˓→excitation using acceleration timeseries

11

12 #To run Uniaxial Inelastic Material, Fiber Section, Nonlinear Mode, MultipleSupport
˓→Earthquake ground motion:First import the InelasticFiberSectionPortal2Dframe.py

13 #(upto gravity loading is already in this script) and run the current script
14 #To run EQ ground-motion analysis (ReadRecord.py, H-E12140.AT2 needs to be downloaded
˓→into the same directory)

15 # MultipleSupport Earthquake ground motion (different acceleration input at specified

˓→support nodes) -- two nodes here

16 #the problem description can be found here:

17 #http://opensees.berkeley.edu/wiki/index.php/Examples_Manual and http://opensees.
˓→berkeley.edu/wiki/index.php/OpenSees_Example_4._Portal_Frame(example: 4)

18 # ------------------------------------------------------------------------------------
˓→--------------

19 # OpenSees (Tcl) code by: Silvia Mazzoni & Frank McKenna, 2006
20 ######################################################################################
˓→####################################################################################

21 import openseespy.opensees as op
22 #import the os module
23 #import os
24 import math
25 op.wipe()
26 ######################################################################################
˓→####################################################################################

27

28 from InelasticFiberSectionPortal2Dframe import *

29 # execute this file after you have built the model, and after you apply gravity
30 #
31

32 # MultipleSupport Earthquake ground motion (different displacement input at spec'd

˓→support nodes) -- two nodes here

33

34 #applying Dynamic Ground motion analysis

35 iSupportNode = [1, 2]
36 iGMfact = [1.5, 1.5]
37 iGMdirection = [1, 1]
38 iGMfile = ['H-E12140', 'H-E12140']
39 DtAnalysis = 0.01*sec # time-step Dt for lateral analysis
40 TmaxAnalysis = 10.0*sec # maximum duration of ground-motion analysis
41 Tol = 1e-8
42

43 # define DAMPING----------------------------------------------------------------------
˓→----------------

44 # apply Rayleigh DAMPING from $xDamp

45 # D=$alphaM*M + $betaKcurr*Kcurrent + $betaKcomm*KlastCommit + $beatKinit*$Kinitial
46 Lambda = op.eigen('-fullGenLapack', 1) # eigenvalue mode 1
47 Omega = math.pow(Lambda, 0.5)
48 betaKcomm = 2 * (0.02/Omega)
49

50 xDamp = 0.02 # 2% damping ratio

(continues on next page)

362 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

51 alphaM = 0.0 # M-prop. damping; D = alphaM*M
52 betaKcurr = 0.0 # K-proportional damping; +beatKcurr*KCurrent
53 betaKinit = 0.0 # initial-stiffness proportional damping +beatKinit*Kini
54

55 op.rayleigh(alphaM,betaKcurr, betaKinit, betaKcomm) # RAYLEIGH damping

56 #-------------------------------------------------------------------------------------
˓→-

57 # --------------------------------- perform Dynamic Ground-Motion Analysis

58 # the following commands are unique to the Multiple-Support Earthquake excitation
59 # Set some parameters
60 IDloadTag = 400 # load tag
61 IDgmSeries = 500 # for multipleSupport Excitation
62

63 # read a PEER strong motion database file, extracts dt from the header and converts
˓→the file

64 # to the format OpenSees expects for Uniform/multiple-support ground motions

65 record = ['H-E12140', 'H-E12140']
66 #dt =[]
67 #nPts = []
68

69 import ReadRecord
70 # Permform the conversion from SMD record to OpenSees record
71 #dt, nPts = ReadRecord.ReadRecord(record+'.at2', record+'.dat')
72 #print(dt, nPts)
73 count = 2
74 #use displacement series, create time series('Path'), then create multi-support
˓→excitation patter (gmtag, 'Plain'), then create imposed ground motion

75 #using groundmotion('nodetag', gmtag), run this in a loop for each support or node
˓→where the earthquake load is going to be applied.

76 op.pattern('MultipleSupport', IDloadTag)
77 for i in range(len(iSupportNode)):
78 record_single = record[i]
79 GMfatt = (iGMfact[i])*g
80 dt, nPts = ReadRecord.ReadRecord(record_single+'.AT2', record_single+'.dat')
81 op.timeSeries('Path', count, '-dt', dt, '-filePath', record_single+'.dat', '-
˓→factor', GMfatt)

82 op.groundMotion(IDgmSeries+count, 'Plain', '-accel', count)

83 op.imposedMotion(iSupportNode[i], iGMdirection[i], IDgmSeries+count)
84 count = count + 1
85

86 maxNumIter = 10
87 op.wipeAnalysis()
88 op.constraints('Transformation')
89 op.numberer('RCM')
90 op.system('BandGeneral')
91 #op.test('EnergyIncr', Tol, maxNumIter)
92 #op.algorithm('ModifiedNewton')
93 #NewmarkGamma = 0.5
94 #NewmarkBeta = 0.25
95 #op.integrator('Newmark', NewmarkGamma, NewmarkBeta)
96 #op.analysis('Transient')
97 #
98 #
99 #Nsteps = int(TmaxAnalysis/ DtAnalysis)
100 #
101 #ok = op.analyze(Nsteps, DtAnalysis)
102
(continues on next page)

1.14. Examples 363

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

103 tCurrent = op.getTime()
104

105 # for gravity analysis, load control is fine, 0.1 is the load factor increment (http:/
˓→/opensees.berkeley.edu/wiki/index.php/Load_Control)

106

107 test = {1:'NormDispIncr', 2: 'RelativeEnergyIncr', 3:'EnergyIncr', 4:

˓→'RelativeNormUnbalance',5: 'RelativeNormDispIncr', 6: 'NormUnbalance'}

108 algorithm = {1:'KrylovNewton', 2: 'SecantNewton' , 3:'ModifiedNewton' , 4:

˓→'RaphsonNewton',5: 'PeriodicNewton', 6: 'BFGS', 7: 'Broyden', 8: 'NewtonLineSearch'}

109

110 #tFinal = TmaxAnalysis

111 tFinal = nPts*dt
112 time = [tCurrent]
113 u3 = [0.0]
114 u4 = [0.0]
115 ok = 0
116

117 while tCurrent < tFinal:

118 # ok = op.analyze(1, .01)
119 for i in test:
120 for j in algorithm:
121 if j < 4:
122 op.algorithm(algorithm[j], '-initial')
123

124 else:
125 op.algorithm(algorithm[j])
126 while ok == 0 and tCurrent < tFinal:
127

128 op.test(test[i], Tol, maxNumIter)

129 NewmarkGamma = 0.5
130 NewmarkBeta = 0.25
131 op.integrator('Newmark', NewmarkGamma, NewmarkBeta)
132 op.analysis('Transient')
133 ok = op.analyze(1, .01)
134

135 if ok == 0 :
136 tCurrent = op.getTime()
137 time.append(tCurrent)
138 u3.append(op.nodeDisp(3,1))
139 u4.append(op.nodeDisp(4,1))
140 print(test[i], algorithm[j], 'tCurrent=', tCurrent)
141

142 import matplotlib.pyplot as plt

143 plt.figure(figsize=(8,8))
144 plt.plot(time, u3)
145 plt.ylabel('Horizontal Displacement of node 3 (in)')
146 plt.xlabel('Time (s)')
147 plt.savefig('Horizontal Disp at Node 3 vs time-multiple support excitation-acctime.
˓→jpeg', dpi = 500)

148 plt.show()
149

150 plt.figure(figsize=(8,8))
151 plt.plot(time, u4)
152 plt.ylabel('Horizontal Displacement of node 4 (in)')
153 plt.xlabel('Time (s)')
154 plt.savefig('Horizontal Disp at Node 4 vs time-multiple support excitation-acctime.
˓→jpeg', dpi = 500)
(continues on next page)

364 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

155 plt.show()
156

157

158 op.wipe()

2D Portal Frame with Units- Multiple Support Dynamic EQ Ground Motion-disptimeseries

Converted to openseespy by: Pavan Chigullapally

University of Auckland
Email: [email protected]

1. To run Uniaxial Inelastic Material, Fiber Section, Nonlinear Mode, MultipleSupport Earthquake ground motion
2. First import the InelasticFiberSectionPortal2Dframe.py
3. To run EQ ground-motion analysis (ReadRecord.py, H-E12140.DT2 needs to be downloaded into the
same directory)
4. MultipleSupport Earthquake ground motion (different displacement input at specified support nodes) – two
nodes here
5. The problem description can be found here (example:4)
6. The source code is shown below, which can be downloaded here.

1 # -*- coding: utf-8 -*-

2 """
3 Created on Mon Apr 22 15:12:06 2019
4

5 @author: pchi893
6 """
7 # Converted to openseespy by: Pavan Chigullapally
8 # University of Auckland
9 # Email: [email protected]
10 # Example4. 2D Portal Frame-- Dynamic EQ input analysis-- multiple-support
˓→excitation using displacement timeseries

11

12 #To run Uniaxial Inelastic Material, Fiber Section, Nonlinear Mode, MultipleSupport
˓→Earthquake ground motion:First import the InelasticFiberSectionPortal2Dframe.py

13 #(upto gravity loading is already in this script) and run the current script
14 #To run EQ ground-motion analysis (ReadRecord.py, H-E12140.DT2 needs to be downloaded
˓→into the same directory)

15 # MultipleSupport Earthquake ground motion (different displacement input at specified

˓→support nodes) -- two nodes here

16 #the problem description can be found here:

17 #http://opensees.berkeley.edu/wiki/index.php/Examples_Manual and http://opensees.
˓→berkeley.edu/wiki/index.php/OpenSees_Example_4._Portal_Frame(example: 4)

18 # ------------------------------------------------------------------------------------
˓→--------------

19 # OpenSees (Tcl) code by: Silvia Mazzoni & Frank McKenna, 2006
20 ######################################################################################
˓→####################################################################################

21 import openseespy.opensees as op
22 #import the os module
23 #import os
24 import math
(continues on next page)

1.14. Examples 365

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

25 op.wipe()
26 ######################################################################################
˓→####################################################################################

27

28 from InelasticFiberSectionPortal2Dframe import *

29 # execute this file after you have built the model, and after you apply gravity
30 #
31

32 # MultipleSupport Earthquake ground motion (different displacement input at spec'd

˓→support nodes) -- two nodes here

33

34 #applying Dynamic Ground motion analysis

35 iSupportNode = [1, 2]
36 iGMfact = [1.5, 1.25]
37 iGMdirection = [1, 1]
38 iGMfile = ['H-E12140', 'H-E12140']
39 DtAnalysis = 0.01*sec # time-step Dt for lateral analysis
40 TmaxAnalysis = 10.0*sec # maximum duration of ground-motion analysis
41 Tol = 1e-8
42

43 # define DAMPING----------------------------------------------------------------------
˓→----------------

44 # apply Rayleigh DAMPING from $xDamp

45 # D=$alphaM*M + $betaKcurr*Kcurrent + $betaKcomm*KlastCommit + $beatKinit*$Kinitial
46 Lambda = op.eigen('-fullGenLapack', 1) # eigenvalue mode 1
47 Omega = math.pow(Lambda, 0.5)
48 betaKcomm = 2 * (0.02/Omega)
49

50 xDamp = 0.02 # 2% damping ratio

51 alphaM = 0.0 # M-prop. damping; D = alphaM*M
52 betaKcurr = 0.0 # K-proportional damping; +beatKcurr*KCurrent
53 betaKinit = 0.0 # initial-stiffness proportional damping +beatKinit*Kini
54

55 op.rayleigh(alphaM,betaKcurr, betaKinit, betaKcomm) # RAYLEIGH damping

56 #-------------------------------------------------------------------------------------
˓→-

57 # --------------------------------- perform Dynamic Ground-Motion Analysis

58 # the following commands are unique to the Multiple-Support Earthquake excitation
59 # Set some parameters
60 IDloadTag = 400 # load tag
61 IDgmSeries = 500 # for multipleSupport Excitation
62

63 # read a PEER strong motion database file, extracts dt from the header and converts
˓→the file

64 # to the format OpenSees expects for Uniform/multiple-support ground motions

65 record = ['H-E12140', 'H-E12140']
66 #dt =[]
67 #nPts = []
68

69 import ReadRecord
70 # Permform the conversion from SMD record to OpenSees record
71 #dt, nPts = ReadRecord.ReadRecord(record+'.at2', record+'.dat')
72 #print(dt, nPts)
73 count = 2
74 #use displacement series, create time series('Path'), then create multi-support
˓→excitation patter (gmtag, 'Plain'), then create imposed ground motion

75 #using groundmotion('nodetag', gmtag), run this in a loop for each support or node
˓→where the earthquake load is going to be applied. (continues on next page)

366 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

76 op.pattern('MultipleSupport', IDloadTag)
77 for i in range(len(iSupportNode)):
78 record_single = record[i]
79 GMfatt = (iGMfact[i])*cm
80 dt, nPts = ReadRecord.ReadRecord(record_single+'.DT2', record_single+'.dat')
81 op.timeSeries('Path', count, '-dt', dt, '-filePath', record_single+'.dat', '-
˓→factor', GMfatt)

82 op.groundMotion(IDgmSeries+count, 'Plain', '-disp', count)

83 op.imposedMotion(iSupportNode[i], iGMdirection[i], IDgmSeries+count)
84 count = count + 1
85

86 maxNumIter = 10
87 op.wipeAnalysis()
88 op.constraints('Transformation')
89 op.numberer('RCM')
90 op.system('BandGeneral')
91 #op.test('EnergyIncr', Tol, maxNumIter)
92 #op.algorithm('ModifiedNewton')
93 #NewmarkGamma = 0.5
94 #NewmarkBeta = 0.25
95 #op.integrator('Newmark', NewmarkGamma, NewmarkBeta)
96 #op.analysis('Transient')
97 #
98 #
99 #Nsteps = int(TmaxAnalysis/ DtAnalysis)
100 #
101 #ok = op.analyze(Nsteps, DtAnalysis)
102

103 tCurrent = op.getTime()

104

105 # for gravity analysis, load control is fine, 0.1 is the load factor increment (http:/
˓→/opensees.berkeley.edu/wiki/index.php/Load_Control)

106

107 test = {1:'NormDispIncr', 2: 'RelativeEnergyIncr', 3:'EnergyIncr', 4:

˓→'RelativeNormUnbalance',5: 'RelativeNormDispIncr', 6: 'NormUnbalance'}

108 algorithm = {1:'KrylovNewton', 2: 'SecantNewton' , 3:'ModifiedNewton' , 4:

˓→'RaphsonNewton',5: 'PeriodicNewton', 6: 'BFGS', 7: 'Broyden', 8: 'NewtonLineSearch'}

109

110 #tFinal = TmaxAnalysis

111 tFinal = nPts*dt
112 time = [tCurrent]
113 u3 = [0.0]
114 u4 = [0.0]
115 ok = 0
116

117 while tCurrent < tFinal:

118 # ok = op.analyze(1, .01)
119 for i in test:
120 for j in algorithm:
121 if j < 4:
122 op.algorithm(algorithm[j], '-initial')
123

124 else:
125 op.algorithm(algorithm[j])
126 while ok == 0 and tCurrent < tFinal:
127

128 op.test(test[i], Tol, maxNumIter)

(continues on next page)

1.14. Examples 367

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

129 NewmarkGamma = 0.5
130 NewmarkBeta = 0.25
131 op.integrator('Newmark', NewmarkGamma, NewmarkBeta)
132 op.analysis('Transient')
133 ok = op.analyze(1, .01)
134

135 if ok == 0 :
136 tCurrent = op.getTime()
137 time.append(tCurrent)
138 u3.append(op.nodeDisp(3,1))
139 u4.append(op.nodeDisp(4,1))
140 print(test[i], algorithm[j], 'tCurrent=', tCurrent)
141

142 import matplotlib.pyplot as plt

143 plt.figure(figsize=(8,8))
144 plt.plot(time, u3)
145 plt.ylabel('Horizontal Displacement of node 3 (in)')
146 plt.xlabel('Time (s)')
147 plt.savefig('Horizontal Disp at Node 3 vs time-multiple support excitation-disptime.
˓→jpeg', dpi = 500)

148 plt.show()
149

150 plt.figure(figsize=(8,8))
151 plt.plot(time, u4)
152 plt.ylabel('Horizontal Displacement of node 4 (in)')
153 plt.xlabel('Time (s)')
154 plt.savefig('Horizontal Disp at Node 4 vs time-multiple support excitation-disptime.
˓→jpeg', dpi = 500)

155 plt.show()
156

157

158 op.wipe()

2D Portal Frame with Units- Uniform Dynamic EQ -bidirectional-acctimeseries

Converted to openseespy by: Pavan Chigullapally

University of Auckland
Email: [email protected]

1. To run Uniaxial Inelastic Material, Fiber Section, Nonlinear Mode, Bidirectional-uniform earthquake ground
motion
2. First import the InelasticFiberSectionPortal2Dframe.py
3. To run EQ ground-motion analysis (ReadRecord.py, H-E12140.AT2, H-E01140.AT2 needs to be
downloaded into the same directory)
4. Bidirectional-uniform support excitation using acceleration timeseries (different accelerations are input at all
support nodes in two directions) – two support nodes here
5. The problem description can be found here (example:4)
6. The source code is shown below, which can be downloaded here.

1 # -*- coding: utf-8 -*-

2 """
(continues on next page)

368 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

3 Created on Mon Apr 22 15:12:06 2019
4

5 @author: pchi893
6 """
7 # Converted to openseespy by: Pavan Chigullapally
8 # University of Auckland
9 # Email: [email protected]
10

11 # Example4. 2D Portal Frame-- Dynamic EQ input analysis-- Bidirectional-uniform

˓→support excitation using acceleration timeseries

12

13 #To run Uniaxial Inelastic Material, Fiber Section, Nonlinear Mode, Bidirectional-
˓→uniform earthquake ground motion:First import the

˓→InelasticFiberSectionPortal2Dframe.py

14 #(upto gravity loading is already in this script) and run the current script
15 #To run EQ ground-motion analysis (ReadRecord.py, H-E12140.AT2 and H-E01140.AT2 needs
˓→to be downloaded into the same directory)

16 # Bidirectional-uniform support excitation using acceleration timeseries (different

˓→accelerations are input at all support nodes in two directions) -- two support

˓→nodes here

17 #the problem description can be found here:

18 #http://opensees.berkeley.edu/wiki/index.php/Examples_Manual and http://opensees.
˓→berkeley.edu/wiki/index.php/OpenSees_Example_4._Portal_Frame(example: 4)

19 # ------------------------------------------------------------------------------------
˓→--------------

20 # OpenSees (Tcl) code by: Silvia Mazzoni & Frank McKenna, 2006
21 # ------------------------------------------------------------------------------------
˓→--------------

22 ######################################################################################
˓→####################################################################################

23 import openseespy.opensees as op
24 #import the os module
25 #import os
26 import math
27 op.wipe()
28 ######################################################################################
˓→####################################################################################

29

30 from InelasticFiberSectionPortal2Dframe import *

31 # execute this file after you have built the model, and after you apply gravity
32 #
33

34 # MultipleSupport Earthquake ground motion (different displacement input at spec'd

˓→support nodes) -- two nodes here

35

36 #applying Dynamic Ground motion analysis

37 #iSupportNode = [1, 2]
38 iGMfact = [1.5, 0.25]
39 iGMdirection = [1, 2]
40 iGMfile = ['H-E01140', 'H-E12140']
41 DtAnalysis = 0.01*sec # time-step Dt for lateral analysis
42 TmaxAnalysis = 10.0*sec # maximum duration of ground-motion analysis
43 Tol = 1e-8
44

45 # define DAMPING----------------------------------------------------------------------
˓→----------------

46 # apply Rayleigh DAMPING from $xDamp

(continues on next page)

1.14. Examples 369

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

47 # D=$alphaM*M + $betaKcurr*Kcurrent + $betaKcomm*KlastCommit + $beatKinit*$Kinitial
48 Lambda = op.eigen('-fullGenLapack', 1) # eigenvalue mode 1
49 Omega = math.pow(Lambda, 0.5)
50 betaKcomm = 2 * (0.02/Omega)
51

52 xDamp = 0.02 # 2% damping ratio

53 alphaM = 0.0 # M-prop. damping; D = alphaM*M
54 betaKcurr = 0.0 # K-proportional damping; +beatKcurr*KCurrent
55 betaKinit = 0.0 # initial-stiffness proportional damping +beatKinit*Kini
56

57 op.rayleigh(alphaM,betaKcurr, betaKinit, betaKcomm) # RAYLEIGH damping

58 #-------------------------------------------------------------------------------------
˓→-

59 # --------------------------------- perform Dynamic Ground-Motion Analysis

60 # the following commands are unique to the Multiple-Support Earthquake excitation
61 # Set some parameters
62 IDloadTag = 400 # load tag
63

64

65 # read a PEER strong motion database file, extracts dt from the header and converts
˓→the file

66 # to the format OpenSees expects for Uniform/multiple-support ground motions

67 record = ['H-E01140', 'H-E12140']
68 #dt =[]
69 #nPts = []
70

71 import ReadRecord
72

73 #this is similar to uniform excitation in single direction

74 count = 2
75 for i in range(len(iGMdirection)):
76 IDloadTag = IDloadTag+count
77 record_single = record[i]
78 GMfatt = (iGMfact[i])*g
79 dt, nPts = ReadRecord.ReadRecord(record_single+'.AT2', record_single+'.dat')
80 op.timeSeries('Path', count, '-dt', dt, '-filePath', record_single+'.dat', '-
˓→factor', GMfatt)

81 op.pattern('UniformExcitation', IDloadTag, iGMdirection[i], '-accel', 2)

82 count = count + 1
83

84 maxNumIter = 10
85 op.wipeAnalysis()
86 op.constraints('Transformation')
87 op.numberer('RCM')
88 op.system('BandGeneral')
89 #op.test('EnergyIncr', Tol, maxNumIter)
90 #op.algorithm('ModifiedNewton')
91 #NewmarkGamma = 0.5
92 #NewmarkBeta = 0.25
93 #op.integrator('Newmark', NewmarkGamma, NewmarkBeta)
94 #op.analysis('Transient')
95

96

97 #Nsteps = int(TmaxAnalysis/ DtAnalysis)

98 #
99 #ok = op.analyze(Nsteps, DtAnalysis)
100
(continues on next page)

370 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

101 tCurrent = op.getTime()
102

103 # for gravity analysis, load control is fine, 0.1 is the load factor increment (http:/
˓→/opensees.berkeley.edu/wiki/index.php/Load_Control)

104

105 test = {1:'NormDispIncr', 2: 'RelativeEnergyIncr', 3:'EnergyIncr', 4:

˓→'RelativeNormUnbalance',5: 'RelativeNormDispIncr', 6: 'NormUnbalance'}

106 algorithm = {1:'KrylovNewton', 2: 'SecantNewton' , 3:'ModifiedNewton' , 4:

˓→'RaphsonNewton',5: 'PeriodicNewton', 6: 'BFGS', 7: 'Broyden', 8: 'NewtonLineSearch'}

107

108 #tFinal = TmaxAnalysis

109 tFinal = nPts*dt
110 time = [tCurrent]
111 u3 = [0.0]
112 u4 = [0.0]
113 ok = 0
114

115 while tCurrent < tFinal:

116 # ok = op.analyze(1, .01)
117 for i in test:
118 for j in algorithm:
119 if j < 4:
120 op.algorithm(algorithm[j], '-initial')
121

122 else:
123 op.algorithm(algorithm[j])
124 while ok == 0 and tCurrent < tFinal:
125

126 op.test(test[i], Tol, maxNumIter)

127 NewmarkGamma = 0.5
128 NewmarkBeta = 0.25
129 op.integrator('Newmark', NewmarkGamma, NewmarkBeta)
130 op.analysis('Transient')
131 ok = op.analyze(1, .01)
132

133 if ok == 0 :
134 tCurrent = op.getTime()
135 time.append(tCurrent)
136 u3.append(op.nodeDisp(3,1))
137 u4.append(op.nodeDisp(4,1))
138 print(test[i], algorithm[j], 'tCurrent=', tCurrent)
139

140 import matplotlib.pyplot as plt

141 plt.figure(figsize=(8,8))
142 plt.plot(time, u3)
143 plt.ylabel('Horizontal Displacement of node 3 (in)')
144 plt.xlabel('Time (s)')
145 plt.savefig('Horizontal Disp at Node 3 vs time-uniform excitation-acctime.jpeg', dpi
˓→= 500)

146 plt.show()
147

148 plt.figure(figsize=(8,8))
149 plt.plot(time, u4)
150 plt.ylabel('Horizontal Displacement of node 4 (in)')
151 plt.xlabel('Time (s)')
152 plt.savefig('Horizontal Disp at Node 4 vs time-uniform excitation-acctime.jpeg', dpi
˓→= 500)
(continues on next page)

1.14. Examples 371

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

153 plt.show()
154 #
155

156 op.wipe()

1.14.3 Tsunami Examples

1. Moving Mesh
2. Background Mesh

Moving Mesh

1. Dambreak Analysis using moving mesh

2. Dambreak with Elastic Obstacle Analysis using moving mesh

Dambreak Analysis using moving mesh

1. The source code is shown below, which can be downloaded here.

2. Run the source code in your favorite Python program.
3. The ParaView is needed to view the results. To view the displaced shape of fluid, use the “Warp By Vector”
filter with scale factor = 1.0.
1 import os
2 import openseespy.opensees as ops
3

4 # ------------------------------
5 # Start of model generation
6 # -----------------------------
7

8 # remove existing model

9 ops.wipe()
10

11 # set modelbuilder
12 ops.model('basic', '-ndm', 2, '-ndf', 2)
13

14 # geometric
15 L = 0.146
16 H = L*2
17 H2 = 0.3
18 h = 0.005
19 alpha = 1.4
20 tw = 3*h
21

22 # material
23 rho = 1000.0
24 mu = 0.0001
25 b1 = 0.0
26 b2 = -9.81
27 thk = 0.012
28 kappa = -1.0
(continues on next page)

372 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

29

30 # time steps
31 dtmax = 1e-3
32 dtmin = 1e-6
33 totaltime = 1.0
34

35 # filename
36 filename = 'dambreak'
37

38 # recorder
39 if not os.path.exists(filename):
40 os.makedirs(filename)
41 ops.recorder('PVD', filename, 'disp', 'vel', 'pressure')
42

43 # nodes
44 ops.node(1, 0.0, 0.0)
45 ops.node(2, L, 0.0)
46 ops.node(3, L, H)
47 ops.node(4, 0.0, H)
48 ops.node(5, 0.0, H2)
49 ops.node(6, 4*L, 0.0)
50 ops.node(7, 4*L, H2)
51 ops.node(8, -tw, H2)
52 ops.node(9, -tw, -tw)
53 ops.node(10, 4*L+tw, -tw)
54 ops.node(11, 4*L+tw, H2)
55

56 # ids for meshing

57 wall_id = 1
58 water_bound_id = -1
59 water_body_id = -2
60

61 # wall mesh
62 wall_tag = 3
63 ndf = 2
64 ops.mesh('line', 1, 9, 4,5,8,9,10,11,7,6,2, wall_id, ndf, h)
65 ops.mesh('line', 2, 3, 2,1,4, wall_id, ndf, h)
66 ops.mesh('tri', wall_tag, 2, 1,2, wall_id, ndf, h)
67

68 # fluid mesh
69 fluid_tag = 4
70 ops.mesh('line', 5, 3, 2,3,4, water_bound_id, ndf, h)
71

72 eleArgs = ['PFEMElementBubble',rho,mu,b1,b2,thk,kappa]
73 ops.mesh('tri', fluid_tag, 2, 2,5, water_body_id, ndf, h, *eleArgs)
74

75 for nd in ops.getNodeTags('-mesh', wall_tag):

76 ops.fix(nd, 1,1)
77

78 # save the original modal

79 ops.record()
80

81 # create constraint object

82 ops.constraints('Plain')
83

84 # create numberer object

85 ops.numberer('Plain')
(continues on next page)

1.14. Examples 373

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

86

87 # create convergence test object

88 ops.test('PFEM', 1e-5, 1e-5, 1e-5, 1e-5, 1e-15, 1e-15, 20, 3, 1, 2)
89

90 # create algorithm object

91 ops.algorithm('Newton')
92

93 # create integrator object

94 ops.integrator('PFEM')
95

96 # create SOE object

97 ops.system('PFEM', '-umfpack', '-print')
98

99 # create analysis object

100 ops.analysis('PFEM', dtmax, dtmin, b2)
101

102 # analysis
103 while ops.getTime() < totaltime:
104

105 # analysis
106 if ops.analyze() < 0:
107 break
108

109 ops.remesh(alpha)
110

111

112

Dambreak with Elastic Obstacle Analysis using moving mesh

1. The source code is shown below, which can be downloaded here.

2. Run the source code in your favorite Python program.
3. The ParaView is needed to view the results. To view the displaced shape of fluid, use the “Warp By Vector”
filter with scale factor = 1.0.

1 import os
2 import openseespy.opensees as ops
3

4 # ------------------------------
5 # Start of model generation
6 # -----------------------------
7

8 # remove existing model

9 ops.wipe()
10

11 # set modelbuilder
12 ops.model('basic', '-ndm', 2, '-ndf', 3)
13

14 # geometric
15 L = 0.146
16 H = 2*L
17 H2 = 0.3
18 b = 0.012
(continues on next page)

374 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

19 h = 0.005
20 alpha = 1.4
21 Hb = 20.0*b/3.0
22 tw = 3*h
23

24 # material
25 rho = 1000.0
26 mu = 0.0001
27 b1 = 0.0
28 b2 = -9.81
29 thk = 0.012
30 kappa = -1.0
31

32 rhos = 2500.0
33 A = thk*thk
34 E = 1e6
35 Iz = thk*thk*thk*thk/12.0
36 bmass = A*Hb*rhos
37

38 # analysis
39 dtmax = 1e-3
40 dtmin = 1e-6
41 totaltime = 1.0
42

43 filename = 'obstacle'
44

45 # recorder
46 if not os.path.exists(filename):
47 os.makedirs(filename)
48 ops.recorder('PVD', filename, 'disp', 'vel', 'pressure')
49

50 # nodes
51 ops.node(1, 0.0, 0.0)
52 ops.node(2, L, 0.0)
53 ops.node(3, L, H, '-ndf', 2)
54 ops.node(4, 0.0, H)
55 ops.node(5, 0.0, H2)
56 ops.node(6, 4*L, 0.0)
57 ops.node(7, 4*L, H2)
58 ops.node(8, -tw, H2)
59 ops.node(9, -tw, -tw)
60 ops.node(10, 4*L+tw, -tw)
61 ops.node(11, 4*L+tw, H2)
62 ops.node(12, 2*L, 0.0)
63 ops.node(13, 2*L, Hb)
64

65 # ids for meshing

66 wall_id = 1
67 beam_id = 2
68 water_bound_id = -1
69 water_body_id = -2
70

71 # transformation
72 transfTag = 1
73 ops.geomTransf('Corotational', transfTag)
74

75 # section
(continues on next page)

1.14. Examples 375

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

76 secTag = 1
77 ops.section('Elastic', secTag, E, A, Iz)
78

79 # beam integration
80 inteTag = 1
81 numpts = 2
82 ops.beamIntegration('Legendre', inteTag, secTag, numpts)
83

84 # beam mesh
85 beamTag = 6
86 ndf = 3
87 ops.mesh('line', beamTag, 2, 12, 13, beam_id, ndf, h, 'dispBeamColumn', transfTag,
˓→inteTag)

88

89 ndmass = bmass/len(ops.getNodeTags('-mesh', beamTag))

90

91 for nd in ops.getNodeTags('-mesh', beamTag):

92 ops.mass(nd, ndmass, ndmass, 0.0)
93

94 # fluid mesh
95 fluidTag = 4
96 ndf = 2
97 ops.mesh('line', 1, 10, 4,5,8,9,10,11,7,6,12,2, wall_id, ndf, h)
98 ops.mesh('line', 2, 3, 2,1,4, wall_id, ndf, h)
99 ops.mesh('line', 3, 3, 2,3,4, water_bound_id, ndf, h)
100

101 eleArgs = ['PFEMElementBubble',rho,mu,b1,b2,thk,kappa]

102 ops.mesh('tri', fluidTag, 2, 2,3, water_body_id, ndf, h, *eleArgs)
103

104 # wall mesh

105 wallTag = 5
106 ops.mesh('tri', wallTag, 2, 1,2, wall_id, ndf, h)
107

108 for nd in ops.getNodeTags('-mesh', wallTag):

109 ops.fix(nd, 1,1,1)
110

111 # save the original modal

112 ops.record()
113

114 # create constraint object

115 ops.constraints('Plain')
116

117 # create numberer object

118 ops.numberer('Plain')
119

120 # create convergence test object

121 ops.test('PFEM', 1e-5, 1e-5, 1e-5, 1e-5, 1e-15, 1e-15, 20, 3, 1, 2)
122

123 # create algorithm object

124 ops.algorithm('Newton')
125

126 # create integrator object

127 ops.integrator('PFEM')
128

129 # create SOE object

130 ops.system('PFEM')
131
(continues on next page)

376 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

132 # create analysis object
133 ops.analysis('PFEM', dtmax, dtmin, b2)
134

135 # analysis
136 while ops.getTime() < totaltime:
137

138 # analysis
139 if ops.analyze() < 0:
140 break
141

142 ops.remesh(alpha)

Background Mesh

1. Dambreak Analysis using background mesh

2. Dambreak with Elastic Obstacle Analysis using background mesh

Dambreak Analysis using background mesh

1. The source code is shown below, which can be downloaded here.

2. Run the source code in your favorite Python program.
3. The ParaView is needed to view the results. To view the displaced shape of fluid, use the “Warp By Vector”
filter with scale factor = 1.0.
1 import os
2 import os.path
3 import openseespy.opensees as ops
4

5 # ------------------------------
6 # Start of model generation
7 # -----------------------------
8

9 # wipe all previous objects

10 ops.wipe()
11

12 # create a model with fluid

13 ops.model('basic', '-ndm', 2, '-ndf', 2)
14

15 # geometric
16 L = 0.146
17 H = L * 2
18 h = L / 40
19

20 # number of particles per cell in each direction

21 numx = 3.0
22 numy = 3.0
23

24 # material
25 rho = 1000.0
26 mu = 0.0001
27 b1 = 0.0
28 b2 = -9.81
(continues on next page)

1.14. Examples 377

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

29 thk = 0.012
30 kappa = -1.0
31

32 # analysis
33 dtmax = 1e-3
34 dtmin = 1e-3
35 totaltime = 1.0
36 filename = 'dambreak-bg'
37

38 # recorder
39 ops.recorder('BgPVD', filename, 'disp', 'vel', 'pressure', '-dT', 1e-3)
40 if not os.path.exists(filename):
41 os.makedirs(filename)
42

43 # fluid particles
44 ndf = 2
45

46 # total number of particles in each direction

47 nx = round(L / h * numx)
48 ny = round(H / h * numy)
49

50 # create particles
51 eleArgs = ['PFEMElementBubble', rho, mu, b1, b2, thk, kappa]
52 partArgs = ['quad', 0.0, 0.0, L, 0.0, L, H, 0.0, H, nx, ny]
53 parttag = 1
54 ops.mesh('part', parttag, *partArgs, *eleArgs, '-vel', 0.0, 0.0)
55

56 print('num particles =', nx * ny)

57

58 # wall
59 ops.node(1, 0.0, H)
60 ops.node(2, 0.0, 0.0)
61 ops.node(3, 4 * L, 0.0)
62 ops.node(4, 4 * L, H)
63

64 walltag = 2
65 wallid = 1
66 ops.mesh('line', walltag, 4, 1, 2, 3, 4, wallid, ndf, h)
67

68 wallnodes = ops.getNodeTags('-mesh', walltag)

69

70 for nd in wallnodes:
71 ops.fix(nd, 1, 1)
72

73 # background mesh
74 lower = [-h, -h]
75 upper = [4 * L + L, H + L]
76

77 ops.mesh('bg', h, *lower, *upper,

78 '-structure', wallid, len(wallnodes), *wallnodes)
79

80 # create constraint object

81 ops.constraints('Plain')
82

83 # create numberer object

84 ops.numberer('Plain')
85
(continues on next page)

378 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

86 # create convergence test object
87 ops.test('PFEM', 1e-5, 1e-5, 1e-5, 1e-5, 1e-5, 1e-5, 10, 3, 1, 2)
88

89 # create algorithm object

90 ops.algorithm('Newton')
91

92 # create integrator object

93 ops.integrator('PFEM', 0.5, 0.25)
94

95 # create SOE object

96 ops.system('PFEM')
97 # ops.system('PFEM', '-mumps) Linux version can use mumps
98

99 # create analysis object

100 ops.analysis('PFEM', dtmax, dtmin, b2)
101

102 # analysis
103 while ops.getTime() < totaltime:
104

105 # analysis
106 if ops.analyze() < 0:
107 break
108

109 ops.remesh()
110

111 print("==========================================")

Dambreak with Elastic Obstacle Analysis using background mesh

1. The source code is shown below, which can be downloaded here.

2. Run the source code in your favorite Python program.
3. The ParaView is needed to view the results. To view the displaced shape of fluid, use the “Warp By Vector”
filter with scale factor = 1.0.

1 import os
2 import openseespy.opensees as ops
3

5 print("=======================================================")
6 print("Starting Dambreak with Obstacle Background Mesh example")
7

8 # ------------------------------
9 # Start of model generation
10 # -----------------------------
11

12 # wipe all previous objects

13 ops.wipe()
14

15 # create a model with fluid

16 ops.model('basic', '-ndm', 2, '-ndf', 3)
17

18 # geometric
19 L = 0.146
(continues on next page)

1.14. Examples 379

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

20 H = L * 2
21 H2 = 0.3
22 b = 0.012
23 h = L / 40
24 Hb = 20.0 * b / 3.0
25

26 # number of particles per cell in each direction

27 numx = 3.0
28 numy = 3.0
29

30 # fluid properties
31 rho = 1000.0
32 mu = 0.0001
33 b1 = 0.0
34 b2 = -9.81
35 thk = 0.012
36 kappa = -1.0
37

38 # elastis structural material

39 rhos = 2500.0
40 A = thk * thk
41 E = 1e6
42 Iz = thk * thk * thk * thk / 12.0
43 bmass = A * Hb * rhos
44

45 # nonlinear structural material

46 E0 = 1e6
47 Fy = 5e4
48 hardening = 0.02
49

50 nonlinear = False
51

52 # analysis
53 dtmax = 1e-3
54 dtmin = 1e-3
55 totaltime = 1.0
56

57 if nonlinear:
58 filename = 'obstaclenonlinear-bg'
59 else:
60 filename = 'obstacle-bg'
61

62 # recorder
63 ops.recorder('BgPVD', filename, 'disp', 'vel', 'pressure', '-dT', 1e-3)
64 if not os.path.exists(filename):
65 os.makedirs(filename)
66

67 # fluid mesh
68 ndf = 3
69

70 # total number of particles in each direction

71 nx = round(L / h * numx)
72 ny = round(H / h * numy)
73

74 # create particles
75 eleArgs = ['PFEMElementBubble', rho, mu, b1, b2, thk, kappa]
76 partArgs = ['quad', 0.0, 0.0, L, 0.0, L, H, 0.0, H, nx, ny]
(continues on next page)

380 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

77 parttag = 1
78 ops.mesh('part', parttag, *partArgs, *eleArgs, '-vel', 0.0, 0.0)
79

80 # wall mesh
81 ops.node(1, 2 * L, 0.0)
82 ops.node(2, 2 * L, Hb)
83 ops.node(3, 0.0, H)
84 ops.node(4, 0.0, 0.0)
85 ops.node(5, 4 * L, 0.0)
86 ops.node(6, 4 * L, H)
87

88 sid = 1
89 walltag = 4
90 ops.mesh('line', walltag, 5, 3, 4, 1, 5, 6, sid, ndf, h)
91

92 wallNodes = ops.getNodeTags('-mesh', walltag)

93 for nd in wallNodes:
94 ops.fix(nd, 1, 1, 1)
95

96 # structural mesh
97

98 # transformation
99 transfTag = 1
100 ops.geomTransf('Corotational', transfTag)
101

102 # section
103 secTag = 1
104 if nonlinear:
105 matTag = 1
106 ops.uniaxialMaterial('Steel01', matTag, Fy, E0, hardening)
107 numfiber = 5
108 ops.section('Fiber', secTag)
109 ops.patch('rect', matTag, numfiber, numfiber, 0.0, 0.0, thk, thk)
110 else:
111 ops.section('Elastic', secTag, E, A, Iz)
112

113 # beam integration

114 inteTag = 1
115 numpts = 2
116 ops.beamIntegration('Legendre', inteTag, secTag, numpts)
117

118 coltag = 3
119 eleArgs = ['dispBeamColumn', transfTag, inteTag]
120 ops.mesh('line', coltag, 2, 1, 2, sid, ndf, h, *eleArgs)
121

122 # mass
123 sNodes = ops.getNodeTags('-mesh', coltag)
124 bmass = bmass / len(sNodes)
125 for nd in sNodes:
126 ops.mass(int(nd), bmass, bmass, 0.0)
127

128

129 # background mesh

130 lower = [-h, -h]
131 upper = [5 * L, 3 * L]
132

133 ops.mesh('bg', h, *lower, *upper,

(continues on next page)

1.14. Examples 381

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

134 '-structure', sid, len(sNodes), *sNodes,
135 '-structure', sid, len(wallNodes), *wallNodes)
136

137 print('num nodes =', len(ops.getNodeTags()))

138 print('num particles =', nx * ny)
139

140 # create constraint object

141 ops.constraints('Plain')
142

143 # create numberer object

144 ops.numberer('Plain')
145

146 # create convergence test object

147 ops.test('PFEM', 1e-5, 1e-5, 1e-5, 1e-5, 1e-5, 1e-5, 100, 3, 1, 2)
148

149 # create algorithm object

150 ops.algorithm('Newton')
151

152 # create integrator object

153 ops.integrator('PFEM', 0.5, 0.25)
154

155 # create SOE object

156 ops.system('PFEM')
157 # system('PFEM', '-mumps') Linux version can use mumps
158

159 # create analysis object

160 ops.analysis('PFEM', dtmax, dtmin, b2)
161

162 # analysis
163 while ops.getTime() < totaltime:
164

165 # analysis
166 if ops.analyze() < 0:
167 break
168

169 ops.remesh()
170

171 print("==========================================")

1.14.4 GeoTechnical Examples

1. Laterally-Loaded Pile Foundation

2. Effective Stress Site Response Analysis of a Layered Soil Column
3. PM4Sand model undrained cyclic simple shear element

Laterally-Loaded Pile Foundation

1. The original model can be found here.

2. The Python code is converted by Pavan Chigullapally from University of Auckland, Auckland
([email protected]), and shown below, which can be downloaded here.

382 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1 # -*- coding: utf-8 -*-

2 """
3 Created on Thu Jan 10 18:24:47 2019
4

5 @author: pchi893
6 """
7

10

11 ##########################################################
12 # #
13 # Procedure to compute ultimate lateral resistance, p_u, #
14 # and displacement at 50% of lateral capacity, y50, for #
15 # p-y springs representing cohesionless soil. #
16 # Converted to openseespy by: Pavan Chigullapally #
17 # University of Auckland #
18 # #
19 # Created by: Hyung-suk Shin #
20 # University of Washington #
21 # Modified by: Chris McGann #
22 # Pedro Arduino #
23 # Peter Mackenzie-Helnwein #
24 # University of Washington #
25 # #
26 ###########################################################
27

28 # references
29 # American Petroleum Institute (API) (1987). Recommended Practice for Planning,
˓→Designing and

30 # Constructing Fixed Offshore Platforms. API Recommended Practice 2A(RP-2A),

˓→Washington D.C,

31 # 17th edition.
32 #
33 # Brinch Hansen, J. (1961). "The ultimate resistance of rigid piles against
˓→transversal forces."

34 # Bulletin No. 12, Geoteknisk Institute, Copenhagen, 59.

35 #
36 # Boulanger, R. W., Kutter, B. L., Brandenberg, S. J., Singh, P., and Chang, D.
˓→(2003). Pile

37 # Foundations in liquefied and laterally spreading ground during earthquakes:

˓→Centrifuge experiments

38 # and analyses. Center for Geotechnical Modeling, University of California at Davis,

˓→ Davis, CA.

39 # Rep. UCD/CGM-03/01.
40 #
41 # Reese, L.C. and Van Impe, W.F. (2001), Single Piles and Pile Groups Under Lateral
˓→Loading.

42 # A.A. Balkema, Rotterdam, Netherlands.

43

44 import math
45

46 def get_pyParam ( pyDepth, gamma, phiDegree, b, pEleLength, puSwitch, kSwitch,

˓→gwtSwitch):

47

48 #----------------------------------------------------------
49 # define ultimate lateral resistance, pult
(continues on next page)

1.14. Examples 383

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

50 #----------------------------------------------------------
51

52 # pult is defined per API recommendations (Reese and Van Impe, 2001 or API, 1987)
˓→ for puSwitch = 1
53 # OR per the method of Brinch Hansen (1961) for puSwitch = 2
54

55 pi = 3.14159265358979
56 phi = phiDegree * (pi/180)
57 zbRatio = pyDepth / b
58

59 #-------API recommended method-------

60

61 if puSwitch == 1:
62

63 # obtain loading-type coefficient A for given depth-to-diameter ratio zb

64 # ---> values are obtained from a figure and are therefore approximate
65 zb = []
66 dataNum = 41
67 for i in range(dataNum):
68 b1 = i * 0.125
69 zb.append(b1)
70 As = [2.8460, 2.7105, 2.6242, 2.5257, 2.4271, 2.3409, 2.2546, 2.1437, 2.0575,
˓→1.9589, 1.8973, 1.8111, 1.7372, 1.6632, 1.5893, 1.5277, 1.4415, 1.3799, 1.3368, 1.

˓→2690, 1.2074, 1.1581,

71 1.1211, 1.0780, 1.0349, 1.0164, 0.9979, 0.9733, 0.9610, 0.9487, 0.9363, 0.

˓→9117, 0.8994, 0.8994, 0.8871, 0.8871, 0.8809, 0.8809, 0.8809, 0.8809, 0.8809]

72

73 # linear interpolation to define A for intermediate values of depth:diameter

˓→ ratio
74 for i in range(dataNum):
75 if zbRatio >= 5.0:
76 A = 0.88
77 elif zb[i] <= zbRatio and zbRatio <= zb[i+1]:
78 A = (As[i+1] - As[i])/(zb[i+1] - zb[i]) * (zbRatio-zb[i]) + As[i]
79

80 # define common terms

81 alpha = phi / 2
82 beta = pi / 4 + phi / 2
83 K0 = 0.4
84

85 tan_1 = math.tan(pi / 4 - phi / 2)

86 Ka = math.pow(tan_1 , 2)
87

88 # terms for Equation (3.44), Reese and Van Impe (2001)

89 tan_2 = math.tan(phi)
90 tan_3 = math.tan(beta - phi)
91 sin_1 = math.sin(beta)
92 cos_1 = math.cos(alpha)
93 c1 = K0 * tan_2 * sin_1 / (tan_3*cos_1)
94

95 tan_4 = math.tan(beta)
96 tan_5 = math.tan(alpha)
97 c2 = (tan_4/tan_3)*tan_4 * tan_5
98

99 c3 = K0 * tan_4 * (tan_2 * sin_1 - tan_5)

100

101 c4 = tan_4 / tan_3 - Ka

(continues on next page)

384 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

102

103 # terms for Equation (3.45), Reese and Van Impe (2001)
104 pow_1 = math.pow(tan_4,8)
105 pow_2 = math.pow(tan_4,4)
106 c5 = Ka * (pow_1-1)
107 c6 = K0 * tan_2 * pow_2
108

109 # Equation (3.44), Reese and Van Impe (2001)

110 pst = gamma * pyDepth * (pyDepth * (c1 + c2 + c3) + b * c4)
111

112 # Equation (3.45), Reese and Van Impe (2001)

113 psd = b * gamma * pyDepth * (c5 + c6)
114

115 # pult is the lesser of pst and psd. At surface, an arbitrary value is defined
116 if pst <=psd:
117 if pyDepth == 0:
118 pu = 0.01
119

120 else:
121 pu = A * pst
122

123 else:
124 pu = A * psd
125

126 # PySimple1 material formulated with pult as a force, not force/length,

˓→ multiply by trib. length
127 pult = pu * pEleLength
128

129 #-------Brinch Hansen method-------

130 elif puSwitch == 2:
131 # pressure at ground surface
132 cos_2 = math.cos(phi)
133

134 tan_6 = math.tan(pi/4+phi/2)

135

136 sin_2 = math.sin(phi)

137 sin_3 = math.sin(pi/4 + phi/2)
138

139 exp_1 = math.exp((pi/2+phi)*tan_2)

140 exp_2 = math.exp(-(pi/2-phi) * tan_2)
141

142 Kqo = exp_1 * cos_2 * tan_6 - exp_2 * cos_2 * tan_1

143 Kco = (1/tan_2) * (exp_1 * cos_2 * tan_6 - 1)
144

145 # pressure at great depth

146 exp_3 = math.exp(pi * tan_2)
147 pow_3 = math.pow(tan_2,4)
148 pow_4 = math.pow(tan_6,2)
149 dcinf = 1.58 + 4.09 * (pow_3)
150 Nc = (1/tan_2)*(exp_3)*(pow_4 - 1)
151 Ko = 1 - sin_2
152 Kcinf = Nc * dcinf
153 Kqinf = Kcinf * Ko * tan_2
154

155 # pressure at an arbitrary depth

156 aq = (Kqo/(Kqinf - Kqo))*(Ko*sin_2/sin_3)
157 KqD = (Kqo + Kqinf * aq * zbRatio)/(1 + aq * zbRatio)
(continues on next page)

1.14. Examples 385

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

158

159 # ultimate lateral resistance

160 if pyDepth == 0:
161 pu = 0.01
162 else:
163 pu = gamma * pyDepth * KqD * b
164

165 # PySimple1 material formulated with pult as a force, not force/length,

˓→ multiply by trib. length
166 pult = pu * pEleLength
167

168 #----------------------------------------------------------
169 # define displacement at 50% lateral capacity, y50
170 #----------------------------------------------------------
171

172 # values of y50 depend of the coefficent of subgrade reaction, k, which can be
˓→defined in several ways.
173 # for gwtSwitch = 1, k reflects soil above the groundwater table
174 # for gwtSwitch = 2, k reflects soil below the groundwater table
175 # a linear variation of k with depth is defined for kSwitch = 1 after API (1987)
176 # a parabolic variation of k with depth is defined for kSwitch = 2 after
˓→Boulanger et al. (2003)

177

178 # API (1987) recommended subgrade modulus for given friction angle, values
˓→ obtained from figure (approximate)
179

180 ph = [28.8, 29.5, 30.0, 31.0, 32.0, 33.0, 34.0, 35.0, 36.0, 37.0, 38.0, 39.0, 40.
˓→ 0]
181

182 # subgrade modulus above the water table

183 if gwtSwitch == 1:
184 k = [10, 23, 45, 61, 80, 100, 120, 140, 160, 182, 215, 250, 275]
185

186 else:
187 k = [10, 20, 33, 42, 50, 60, 70, 85, 95, 107, 122, 141, 155]
188

189 dataNum = 13
190 for i in range(dataNum):
191 if ph[i] <= phiDegree and phiDegree <= ph[i+1]:
192 khat = (k[i+1]-k[i])/(ph[i+1]-ph[i])*(phiDegree - ph[i]) + k[i]
193

194 # change units from (lb/in^3) to (kN/m^3)

195 k_SIunits = khat * 271.45
196

197 # define parabolic distribution of k with depth if desired (i.e. lin_par switch
˓→ == 2)
198 sigV = pyDepth * gamma
199

200 if sigV == 0:
201 sigV = 0.01
202

203 if kSwitch == 2:
204 # Equation (5-16), Boulanger et al. (2003)
205 cSigma = math.pow(50 / sigV , 0.5)
206 # Equation (5-15), Boulanger et al. (2003)
207 k_SIunits = cSigma * k_SIunits
208
(continues on next page)

386 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

209 # define y50 based on pult and subgrade modulus k
210

211 # based on API (1987) recommendations, p-y curves are described using tanh
˓→ functions.
212 # tcl does not have the atanh function, so must define this specifically
213

214 # i.e. atanh(x) = 1/2*ln((1+x)/(1-x)), |x| < 1

215

216 # when half of full resistance has been mobilized, p(y50)/pult = 0.5
217 x = 0.5
218 log_1 = math.log((1+x)/(1-x))
219 atanh_value = 0.5 * log_1
220

221 # need to be careful at ground surface (don't want to divide by zero)

222 if pyDepth == 0.0:
223 pyDepth = 0.01
224

225 y50 = 0.5 * (pu/ A)/(k_SIunits * pyDepth) * atanh_value

226 # return pult and y50 parameters
227 outResult = []
228 outResult.append(pult)
229 outResult.append(y50)
230

231 return outResult

232

233 ######################################################################################
˓→###################################################################################

234

235 ######################################################################################
˓→###################################################################################

236

237 ###########################################################
238 # #
239 # Procedure to compute ultimate tip resistance, qult, and #
240 # displacement at 50% mobilization of qult, z50, for #
241 # use in q-z curves for cohesionless soil. #
242 # Converted to openseespy by: Pavan Chigullapally #
243 # University of Auckland #
244 # Created by: Chris McGann #
245 # Pedro Arduino #
246 # University of Washington #
247 # #
248 ###########################################################
249

250 # references
251 # Meyerhof G.G. (1976). "Bearing capacity and settlement of pile foundations."
252 # J. Geotech. Eng. Div., ASCE, 102(3), 195-228.
253 #
254 # Vijayvergiya, V.N. (1977). "Load-movement characteristics of piles."
255 # Proc., Ports 77 Conf., ASCE, New York.
256 #
257 # Kulhawy, F.H. ad Mayne, P.W. (1990). Manual on Estimating Soil Properties for
258 # Foundation Design. Electrical Power Research Institute. EPRI EL-6800,
259 # Project 1493-6 Final Report.
260

261 def get_qzParam (phiDegree, b, sigV, G):

262
(continues on next page)

1.14. Examples 387

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

263 # define required constants; pi, atmospheric pressure (kPa), pa, and coeff. of
˓→lat earth pressure, Ko

264 pi = 3.14159265358979
265 pa = 101
266 sin_4 = math.sin(phiDegree * (pi/180))
267 Ko = 1 - sin_4
268

269 # ultimate tip pressure can be computed by qult = Nq*sigV after Meyerhof (1976)
270 # where Nq is a bearing capacity factor, phi is friction angle, and sigV is eff.
˓→overburden

271 # stress at the pile tip.

272 phi = phiDegree * (pi/180)
273

274 # rigidity index

275 tan_7 = math.tan(phi)
276 Ir = G/(sigV * tan_7)
277 # bearing capacity factor
278 tan_8 = math.tan(pi/4+phi/2)
279 sin_5 = math.sin(phi)
280 pow_4 = math.pow(tan_8,2)
281 pow_5 = math.pow(Ir,(4*sin_5)/(3*(1+sin_5)))
282 exp_4 = math.exp(pi/2-phi)
283

284 Nq = (1+2*Ko)*(1/(3-sin_5))*exp_4*(pow_4)*(pow_5)
285 # tip resistance
286 qu = Nq * sigV
287 # QzSimple1 material formulated with qult as force, not stress, multiply by area of
˓→pile tip

288 pow_6 = math.pow(b, 2)

289 qult = qu * pi*pow_6/4
290

291 # the q-z curve of Vijayvergiya (1977) has the form, q(z) = qult*(z/zc)^(1/3)
292 # where zc is critical tip deflection given as ranging from 3-9% of the
293 # pile diameter at the tip.
294

295 # assume zc is 5% of pile diameter

296 zc = 0.05 * b
297

298 # based on Vijayvergiya (1977) curve, z50 = 0.125*zc

299 z50 = 0.125 * zc
300

301 # return values of qult and z50 for use in q-z material
302 outResult = []
303 outResult.append(qult)
304 outResult.append(z50)
305

306 return outResult

307

308 ######################################################################################
˓→###################################################################################

309

310 ######################################################################################
˓→###################################################################################

311 ##########################################################
312 # #
313 # Procedure to compute ultimate resistance, tult, and #
314 # displacement at 50% mobilization of tult, z50, for #
(continues on next page)

388 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

315 # use in t-z curves for cohesionless soil. #
316 # Converted to openseespy by: Pavan Chigullapally #
317 # University of Auckland #
318 # Created by: Chris McGann #
319 # University of Washington #
320 # #
321 ###########################################################
322

323 def get_tzParam ( phi, b, sigV, pEleLength):

324

325 # references
326 # Mosher, R.L. (1984). "Load transfer criteria for numerical analysis of
327 # axial loaded piles in sand." U.S. Army Engineering and Waterways
328 # Experimental Station, Automatic Data Processing Center, Vicksburg, Miss.
329 #
330 # Kulhawy, F.H. (1991). "Drilled shaft foundations." Foundation engineering
331 # handbook, 2nd Ed., Chap 14, H.-Y. Fang ed., Van Nostrand Reinhold, New York
332

333 pi = 3.14159265358979
334

335 # Compute tult based on tult = Ko*sigV*pi*dia*tan(delta), where

336 # Ko is coeff. of lateral earth pressure at rest,
337 # taken as Ko = 0.4
338 # delta is interface friction between soil and pile,
339 # taken as delta = 0.8*phi to be representative of a
340 # smooth precast concrete pile after Kulhawy (1991)
341

342 delta = 0.8 * phi * pi/180

343

344 # if z = 0 (ground surface) need to specify a small non-zero value of sigV

345

346 if sigV == 0.0:

347 sigV = 0.01
348

349 tan_9 = math.tan(delta)

350 tu = 0.4 * sigV * pi * b * tan_9
351

352 # TzSimple1 material formulated with tult as force, not stress, multiply by
˓→ tributary length of pile
353 tult = tu * pEleLength
354

355 # Mosher (1984) provides recommended initial tangents based on friction angle
356 # values are in units of psf/in
357 kf = [6000, 10000, 10000, 14000, 14000, 18000]
358 fric = [28, 31, 32, 34, 35, 38]
359

360 dataNum = len(fric)

361

362

363 # determine kf for input value of phi, linear interpolation for intermediate
˓→ values
364 if phi < fric[0]:
365 k = kf[0]
366 elif phi > fric[5]:
367 k = kf[5]
368 else:
369 for i in range(dataNum):
(continues on next page)

1.14. Examples 389

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

370 if fric[i] <= phi and phi <= fric[i+1]:
371 k = ((kf[i+1] - kf[i])/(fric[i+1] - fric[i])) * (phi - fric[i]) +
˓→ kf[i]
372

373

374 # need to convert kf to units of kN/m^3

375 kSIunits = k * 1.885
376

377 # based on a t-z curve of the shape recommended by Mosher (1984), z50 = tult/kf
378 z50 = tult / kSIunits
379

380 # return values of tult and z50 for use in t-z material
381 outResult = []
382 outResult.append(tult)
383 outResult.append(z50)
384

385 return outResult

386

387

388 ######################################################################################
˓→###################################################################################

389

390 ######################################################################################
˓→###################################################################################

391

392 ###########################################################
393 # #
394 # Static pushover of a single pile, modeled as a beam on #
395 # a nonlinear Winkler foundation. Lateral soil response #
396 # is described by p-y springs. Vertical soil response #
397 # described by t-z and q-z springs. #
398 # Converted to openseespy by: Pavan Chigullapally #
399 # University of Auckland #
400 # Created by: Chris McGann #
401 # HyungSuk Shin #
402 # Pedro Arduino #
403 # Peter Mackenzie-Helnwein #
404 # --University of Washington-- #
405 # #
406 # ---> Basic units are kN and meters #
407 # #
408 ###########################################################
409

410

411 import openseespy.opensees as op

412

413 op.wipe()
414

415 ######################################################################################
˓→###################################################################################

416

417 ######################################################################################
˓→###################################################################################

418

419 # all the units are in SI units N and mm

420

421 #----------------------------------------------------------
(continues on next page)

390 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

422 # pile geometry and mesh
423 #----------------------------------------------------------
424

425 # length of pile head (above ground surface) (m)

426 L1 = 1.0
427 # length of embedded pile (below ground surface) (m)
428 L2 = 20.0
429 # pile diameter
430 diameter = 1.0
431

432 # number of pile elements

433 nElePile = 84
434 # pile element length
435 eleSize = (L1+L2)/nElePile
436

437 # number of total pile nodes

438 nNodePile = 1 + nElePile
439

440 #----------------------------------------------------------
441 # create spring nodes
442 #----------------------------------------------------------
443 # spring nodes created with 3 dim, 3 dof
444 op.model('basic', '-ndm', 3, '-ndf', 3)
445

446 # counter to determine number of embedded nodes

447 count = 0
448

449 # create spring nodes

450

451 #1 to 85 are spring nodes

452

453 pile_nodes = dict()

454

455 for i in range(nNodePile):

456 zCoord = eleSize * i
457 if zCoord <= L2:
458 op.node(i+1, 0.0, 0.0, zCoord)
459 op.node(i+101, 0.0, 0.0, zCoord)
460 pile_nodes[i+1] = (0.0, 0.0, zCoord)
461 pile_nodes[i+101] = (0.0, 0.0, zCoord)
462 count = count + 1
463

464 print("Finished creating all spring nodes...")

465

466 # number of embedded nodes

467 nNodeEmbed = count
468

469 # spring node fixities

470 for i in range(nNodeEmbed):
471 op.fix(i+1, 1, 1, 1)
472 op.fix(i+101, 0, 1, 1)
473

474 print("Finished creating all spring node fixities...")

475

476 #----------------------------------------------------------
477 # soil properties
478 #----------------------------------------------------------
(continues on next page)

1.14. Examples 391

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

479

480 # soil unit weight (kN/m^3)

481 gamma = 17.0
482 # soil internal friction angle (degrees)
483 phi = 36.0
484 # soil shear modulus at pile tip (kPa)
485 Gsoil = 150000.0
486

487 # select pult definition method for p-y curves

488 # API (default) --> 1
489 # Brinch Hansen --> 2
490 puSwitch = 1
491

492 # variation in coefficent of subgrade reaction with depth for p-y curves
493 # API linear variation (default) --> 1
494 # modified API parabolic variation --> 2
495 kSwitch = 1
496

497 # effect of ground water on subgrade reaction modulus for p-y curves
498 # above gwt --> 1
499 # below gwt --> 2
500 gwtSwitch = 1
501

502 #----------------------------------------------------------
503 # create spring material objects
504 #----------------------------------------------------------
505

506 # p-y spring material

507

508 for i in range(1 , nNodeEmbed+1):

509 # depth of current py node
510 pyDepth = L2 - eleSize * (i-1)
511 # procedure to define pult and y50
512 pyParam = get_pyParam(pyDepth, gamma, phi, diameter, eleSize, puSwitch, kSwitch,
˓→gwtSwitch)

513 pult = pyParam [0]

514 y50 = pyParam [1]
515 op.uniaxialMaterial('PySimple1', i, 2, pult, y50, 0.0)
516

517

518 # t-z spring material

519 for i in range(2, nNodeEmbed+1):
520 # depth of current tz node
521 pyDepth = eleSize * (i-1)
522 # vertical effective stress at current depth
523 sigV = gamma * pyDepth
524 # procedure to define tult and z50
525 tzParam = get_tzParam(phi, diameter, sigV, eleSize)
526 tult = tzParam [0]
527 z50 = tzParam [1]
528 op.uniaxialMaterial('TzSimple1', i+100, 2, tult, z50, 0.0)
529

530

531 # q-z spring material

532

533 # vertical effective stress at pile tip, no water table (depth is embedded pile
˓→ length)
(continues on next page)

392 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

534 sigVq = gamma * L2
535 # procedure to define qult and z50
536 qzParam = get_qzParam (phi, diameter, sigVq, Gsoil)
537 qult = qzParam [0]
538 z50q = qzParam [1]
539

540 #op.uniaxialMaterial('QzSimple1', 101, 2, qult, z50q) #, 0.0, 0.0

541 op.uniaxialMaterial('TzSimple1', 101, 2, qult, z50q, 0.0)
542

543 print("Finished creating all p-y, t-z, and z-z spring material objects...")
544

545

546 #----------------------------------------------------------
547 # create zero-length elements for springs
548 #----------------------------------------------------------
549

550 # element at the pile tip (has q-z spring)

551 op.element('zeroLength', 1001, 1, 101, '-mat', 1, 101, '-dir', 1, 3)
552

553 # remaining elements

554 for i in range(2, nNodeEmbed+1):
555 op.element('zeroLength', 1000+i, i, 100+i, '-mat', i, 100+i, '-dir', 1, 3)
556

557 print("Finished creating all zero-Length elements for springs...")

558

559 #----------------------------------------------------------
560 # create pile nodes
561 #----------------------------------------------------------
562

563 # pile nodes created with 3 dimensions, 6 degrees of freedom

564 op.model('basic', '-ndm', 3, '-ndf', 6)
565

566 # create pile nodes

567 for i in range(1, nNodePile+1):
568 zCoord = eleSize * i
569 op.node(i+200, 0.0, 0.0, zCoord)
570

571 print("Finished creating all pile nodes...")

572

573 # create coordinate-transformation object

574 op.geomTransf('Linear', 1, 0.0, -1.0, 0.0)
575

576

577 # create fixity at pile head (location of loading)

578 op.fix(200+nNodePile, 0, 1, 0, 1, 0, 1)
579

580

581 # create fixities for remaining pile nodes

582 for i in range(201, 200+nNodePile):
583 op.fix(i, 0, 1, 0, 1, 0, 1)
584

585 print("Finished creating all pile node fixities...")

586

587 #----------------------------------------------------------
588 # define equal dof between pile and spring nodes
589 #----------------------------------------------------------
590
(continues on next page)

1.14. Examples 393

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

591 for i in range(1, nNodeEmbed+1):
592 op.equalDOF(200+i, 100+i, 1, 3)
593

594 print("Finished creating all equal degrees of freedom...")

595

596 #----------------------------------------------------------
597 # pile section
598 #----------------------------------------------------------
599 ######################################################################################
˓→####################################################################################

600

601 ######################################################################################
˓→###################################################################################

602

603 #----------------------------------------------------------
604 # create elastic pile section
605 #----------------------------------------------------------
606

607 secTag = 1
608 E = 25000000.0
609 A = 0.785
610 Iz = 0.049
611 Iy = 0.049
612 G = 9615385.0
613 J = 0.098
614

615 matTag = 3000

616 op.section('Elastic', 1, E, A, Iz, Iy, G, J)
617

618 # elastic torsional material for combined 3D section

619 op.uniaxialMaterial('Elastic', 3000, 1e10)
620

621 # create combined 3D section

622 secTag3D = 3
623 op.section('Aggregator', secTag3D, 3000, 'T', '-section', 1)
624

625

626 ######################################################################################
˓→###################################################################################

627

628 ######################################################################################
˓→####################################################################################

629

630 # elastic pile section

631 #import elasticPileSection
632

633 #----------------------------------------------------------
634 # create pile elements
635 #----------------------------------------------------------
636 op.beamIntegration('Legendre', 1, secTag3D, 3) # we are using gauss-Legendre
˓→integration as it is the default integration scheme used in opensees tcl (check

˓→dispBeamColumn)

637

638 for i in range(201, 201+nElePile):

639 op.element('dispBeamColumn', i, i, i+1, 1, 1)
640

641 print("Finished creating all pile elements...")

(continues on next page)

394 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

642

643 #----------------------------------------------------------
644 # create recorders
645 #----------------------------------------------------------
646

647 # record information at specified increments

648 timeStep = 0.5
649

650 # record displacements at pile nodes

651 op.recorder('Node', '-file', 'pileDisp.out','-time', '-dT', timeStep, '-nodeRange',
˓→201, 200 + nNodePile, '-dof', 1,2,3, 'disp')

652

653 # record reaction force in the p-y springs

654 op.recorder('Node', '-file', 'reaction.out','-time', '-dT', timeStep, '-nodeRange', 1,
˓→ nNodePile, '-dof', 1, 'reaction')

655

656 # record element forces in pile elements

657 op.recorder('Element', '-file', 'pileForce.out','-time', '-dT', timeStep, '-eleRange',
˓→ 201, 200+nElePile, 'globalForce')

658

659 print("Finished creating all recorders...")

660

661 #----------------------------------------------------------
662 # create the loading
663 #----------------------------------------------------------
664

665 op.setTime(10.0)
666

667 # apply point load at the uppermost pile node in the x-direction
668 values = [0.0, 0.0, 1.0, 1.0]
669 time = [0.0, 10.0, 20.0, 10000.0]
670

671 nodeTag = 200+nNodePile

672 loadValues = [3500.0, 0.0, 0.0, 0.0, 0.0, 0.0]
673 op.timeSeries('Path', 1, '-values', *values, '-time', *time, '-factor', 1.0)
674

675 op.pattern('Plain', 10, 1)

676 op.load(nodeTag, *loadValues)
677

678 print("Finished creating loading object...")

679

680 #----------------------------------------------------------
681 # create the analysis
682 #----------------------------------------------------------
683 op.integrator('LoadControl', 0.05)
684 op.numberer('RCM')
685 op.system('SparseGeneral')
686 op.constraints('Transformation')
687 op.test('NormDispIncr', 1e-5, 20, 1)
688 op.algorithm('Newton')
689 op.analysis('Static')
690

691 print("Starting Load Application...")

692 op.analyze(201)
693

694 print("Load Application finished...")

695 #print("Loading Analysis execution time: [expr $endT-$startT] seconds.")
(continues on next page)

1.14. Examples 395

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

696

697 #op.wipe
698

699 op.reactions()
700 Nodereactions = dict()
701 Nodedisplacements = dict()
702 for i in range(201,nodeTag+1):
703 Nodereactions[i] = op.nodeReaction(i)
704 Nodedisplacements[i] = op.nodeDisp(i)
705 print('Node Reactions are: ', Nodereactions)
706 print('Node Displacements are: ', Nodedisplacements)
707

708

709

710

711

Effective Stress Site Response Analysis of a Layered Soil Column

1. The original model can be found here.

2. The Python code is converted by Harsh Mistry from The University of Manchester, UK
([email protected]), and shown below, which can be downloaded here.
3. The Input motion (velocity-time) can be downloaded here.
1 # -*- coding: utf-8 -*-
2 """
3 Created on Fri Jan 29 16:39:41 2021
4

5 @author: harsh
6 """
7 import numpy as np
8 import math as mm
9 import opensees as op
10 import time as tt
11 ##################################################################
12 # #
13 # Effective stress site response analysis for a layered #
14 # soil profile located on a 2% slope and underlain by an #
15 # elastic half-space. 9-node quadUP elements are used. #
16 # The finite rigidity of the elastic half space is #
17 # considered through the use of a viscous damper at the #
18 # base. #
19 # #
20 # Converted to openseespy by: Harsh Mistry #
21 # The University of Manchester #
22 # #
23 # Created by: Chris McGann #
24 # HyungSuk Shin #
25 # Pedro Arduino #
26 # Peter Mackenzie-Helnwein #
27 # --University of Washington-- #
28 # #
29 # ---> Basic units are kN and m (unless specified) #
30 # #
(continues on next page)

396 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

31 ##################################################################
32 #-------------------------------------------------------------------------------------
˓→----

33 # 1. DEFINE SOIL AND MESH GEOMETRY

34 #-------------------------------------------------------------------------------------
˓→----

35

36 op.wipe()
37 nodes_dict = dict()
38

39 #---SOIL GEOMETRY
40 # thicknesses of soil profile (m)
41 soilThick = 30.0
42 # number of soil layers
43 numLayers = 3
44 # layer thicknesses
45 layerThick=[20.0,8.0,2.0]
46

47 # depth of water table

48 waterTable = 2.0
49

50 # define layer boundaries

51 layerBound=np.zeros((numLayers,1))
52 layerBound[0]=layerThick[0];
53 for i in range(1,numLayers):
54 layerBound[i]=layerBound[i-1]+layerThick[i]
55

56 #---MESH GEOMETRY
57 # number of elements in horizontal direction
58 nElemX = 1
59 # number of nodes in horizontal direction
60 nNodeX =2 * nElemX+1
61 # horizontal element size (m)
62 sElemX = 2.0
63

64 # number of elements in vertical direction for each layer

65 nElemY = [40,16,4]
66

67 # total number of elements in vertical direction

68 nElemT = 60
69

70 sElemY = np.zeros((numLayers,1))
71 # vertical element size in each layer
72 for i in range(numLayers):
73 sElemY[i] = [layerThick[i-1]/nElemY[i-1]]
74 print('size:',sElemY[i])
75

76 # number of nodes in vertical direction

77 nNodeY = 2 * nElemT+1
78

79 # total number of nodes

80 nNodeT = nNodeX * nNodeY
81

82 #-------------------------------------------------------------------------------------
˓→----

83 # 2. CREATE PORE PRESSURE NODES AND FIXITIES

84 #-------------------------------------------------------------------------------------
˓→---- (continues on next page)

1.14. Examples 397

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

85 op.model('basic', '-ndm', 2, '-ndf', 3)
86

87 count = 1
88 layerNodeCount = 0
89 dry_Node=np.zeros((500,1))
90 node_save=np.zeros((500,1))
91 # loop over soil layers
92 for k in range(1,numLayers+1):
93 # loop in horizontal direction
94 for i in range(1,nNodeX+1,2):
95 if k==1:
96 bump = 1
97 else:
98 bump = 0
99 j_end=2 * nElemY[k-1] + bump
100 for j in range(1,j_end+1,2):
101 xCoord = (i-1) * (sElemX/2)
102 yctr = j + layerNodeCount
103 yCoord = (yctr-1) * (np.float(sElemY[k-1]))/2
104 nodeNum = i + ((yctr-1) * nNodeX)
105 op.node(nodeNum, xCoord, yCoord)
106

107 # output nodal information to data file

108 nodes_dict[nodeNum] = (nodeNum, xCoord, yCoord)
109 node_save[nodeNum] = np.int(nodeNum)
110 # designate nodes above water table
111 waterHeight = soilThick - waterTable
112 if yCoord >= waterHeight:
113 dry_Node[count] = nodeNum
114 count = count+1
115 layerNodeCount = yctr + 1
116

117 dryNode=np.trim_zeros(dry_Node)
118 Node_d=np.unique(node_save)
119 Node_d=np.trim_zeros(Node_d)
120 np.savetxt('Node_record.txt',Node_d)
121 print('Finished creating all -ndf 3 nodes')
122 print('Number of Dry Nodes:',len(dryNode))
123

124 # define fixities for pore pressure nodes above water table
125 for i in range(count-1):
126 n_dryNode=np.int(dryNode[i])
127 op.fix(n_dryNode, 0, 0, 1)
128

129 op.fix(1, 0, 1, 0)
130 op.fix(3, 0, 1, 0)
131 print('Finished creating all -ndf 3 boundary conditions...')
132

133 # define equal degrees of freedom for pore pressure nodes

134 for i in range(1,((3*nNodeY)-2),6):
135 op.equalDOF(i, i+2, 1, 2)
136

137 print("Finished creating equalDOF for pore pressure nodes...")

138

139 #-------------------------------------------------------------------------------------
˓→----

140 # 3. CREATE INTERIOR NODES AND FIXITIES

(continues on next page)

398 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

141 #-------------------------------------------------------------------------------------
˓→----

142 op.model('basic', '-ndm', 2, '-ndf', 2)

143

144 xCoord = np.float(sElemX/2)

145

146 # loop over soil layers

147 layerNodeCount = 0
148

149 for k in range(1,numLayers+1):

150 if k==1:
151 bump = 1
152 else:
153 bump = 0
154 j_end=2 * nElemY[k-1] + bump
155 for j in range(1,j_end+1,1):
156 yctr = j + layerNodeCount
157 yCoord = (yctr-1) * (np.float(sElemY[k-1]))/2
158 nodeNum = (3*yctr) - 1
159 op.node(nodeNum, xCoord, yCoord)
160 # output nodal information to data file
161 nodes_dict[nodeNum] = (nodeNum, xCoord, yCoord)
162

163 layerNodeCount = yctr

164

165 # interior nodes on the element edges

166 # loop over layers
167 layerNodeCount = 0
168

169 for k in range(1,numLayers+1):

170 # loop in vertical direction
171 for j in range(1,((nElemY[k-1])+1)):
172 yctr = j + layerNodeCount;
173 yCoord = np.float(sElemY[k-1])*(yctr-0.5)
174 nodeNumL = (6*yctr) - 2
175 nodeNumR = nodeNumL + 2
176

177 op.node(nodeNumL ,0.0, yCoord)

178 op.node(nodeNumR , sElemX, yCoord)
179

180 # output nodal information to data file

181 nodes_dict[nodeNumL] = (nodeNumL ,0.0, yCoord)
182 nodes_dict[nodeNumR] = (nodeNumR , sElemX, yCoord)
183 layerNodeCount = yctr
184

185 print("Finished creating all -ndf 2 nodes...")

186

187 # define fixities for interior nodes at base of soil column

188 op.fix(2, 0, 1)
189 print('Finished creating all -ndf 2 boundary conditions...')
190

191 # define equal degrees of freedom which have not yet been defined
192 for i in range(1,((3*nNodeY)-6),6):
193 op.equalDOF(i , i+1, 1, 2)
194 op.equalDOF(i+3, i+4, 1, 2)
195 op.equalDOF(i+3, i+5, 1, 2)
196
(continues on next page)

1.14. Examples 399

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

197 op.equalDOF(nNodeT-2, nNodeT-1, 1, 2)
198 print('Finished creating equalDOF constraints...')
199

200 #-------------------------------------------------------------------------------------
˓→----

201 # 4. CREATE SOIL MATERIALS

202 #-------------------------------------------------------------------------------------
˓→----

203

204 # define grade of slope (%)

205 grade = 2.0
206 slope = mm.atan(grade/100.0)
207 g = -9.81
208

209 xwgt_var = g * (mm.sin(slope))

210 ywgt_var = g * (mm.cos(slope))
211 thick = [1.0,1.0,1.0]
212 xWgt = [xwgt_var, xwgt_var, xwgt_var]
213 yWgt = [ywgt_var, ywgt_var, ywgt_var]
214 uBulk = [6.88E6, 5.06E6, 5.0E-6]
215 hPerm = [1.0E-4, 1.0E-4, 1.0E-4]
216 vPerm = [1.0E-4, 1.0E-4, 1.0E-4]
217

218

219 # nDMaterial PressureDependMultiYield02

220 # nDMaterial('PressureDependMultiYield02', matTag, nd, rho, refShearModul,
˓→refBulkModul,\

221 # frictionAng, peakShearStra, refPress, pressDependCoe, PTAng,\

222 # contrac[0], contrac[2], dilat[0], dilat[2], noYieldSurf=20.0,\
223 # *yieldSurf=[], contrac[1]=5.0, dilat[1]=3.0, *liquefac=[1.0,0.0],e=0.6, \
224 # *params=[0.9, 0.02, 0.7, 101.0], c=0.1)
225

226 op.nDMaterial('PressureDependMultiYield02',3, 2, 1.8, 9.0e4, 2.2e5, 32, 0.1, \

227 101.0, 0.5, 26, 0.067, 0.23, 0.06, \
228 0.27, 20, 5.0, 3.0, 1.0, \
229 0.0, 0.77, 0.9, 0.02, 0.7, 101.0)
230

231 op.nDMaterial('PressureDependMultiYield02', 2, 2, 2.24, 9.0e4, 2.2e5, 32, 0.1, \

232 101.0, 0.5, 26, 0.067, 0.23, 0.06, \
233 0.27, 20, 5.0, 3.0, 1.0, \
234 0.0, 0.77, 0.9, 0.02, 0.7, 101.0)
235

236 op.nDMaterial('PressureDependMultiYield02',1, 2, 2.45, 1.3e5, 2.6e5, 39, 0.1, \

237 101.0, 0.5, 26, 0.010, 0.0, 0.35, \
238 0.0, 20, 5.0, 3.0, 1.0, \
239 0.0, 0.47, 0.9, 0.02, 0.7, 101.0)
240

241 print("Finished creating all soil materials...")

242

243 #-------------------------------------------------------------------------------------
˓→----

244 # 5. CREATE SOIL ELEMENTS

245 #-------------------------------------------------------------------------------------
˓→----

246

247 for j in range(1,nElemT+1):

248 nI = ( 6*j) - 5
(continues on next page)

400 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

249 nJ = nI + 2
250 nK = nI + 8
251 nL = nI + 6
252 nM = nI + 1
253 nN = nI + 5
254 nP = nI + 7
255 nQ = nI + 3
256 nR = nI + 4
257

258 lowerBound = 0.0

259 for i in range(1,numLayers+1):
260 if j * sElemY[i-1] <= layerBound[i-1] and j * sElemY[i-1] > lowerBound:
261 # permeabilities are initially set at 1.0 m/s for gravity analysis,
262 op.element('9_4_QuadUP', j, nI, nJ, nK, nL, nM, nN, nP, nQ, nR, \
263 thick[i-1], i, uBulk[i-1], 1.0, 1.0, 1.0, xWgt[i-1],
˓→yWgt[i-1])

264

265 lowerBound = layerBound[i-1]

266

267 print("Finished creating all soil elements...")

268 #-------------------------------------------------------------------------------------
˓→----

269 # 6. LYSMER DASHPOT

270 #-------------------------------------------------------------------------------------
˓→----

271

272 # define dashpot nodes

273 dashF = nNodeT+1
274 dashS = nNodeT+2
275

276 op.node(dashF, 0.0, 0.0)

277 op.node(dashS, 0.0, 0.0)
278

279 # define fixities for dashpot nodes

280 op.fix(dashF, 1, 1)
281 op.fix(dashS, 0, 1)
282

283 # define equal DOF for dashpot and base soil node
284 op.equalDOF(1, dashS, 1)
285 print('Finished creating dashpot nodes and boundary conditions...')
286

287 # define dashpot material

288 colArea = sElemX * thick[0]
289 rockVS = 700.0
290 rockDen = 2.5
291 dashpotCoeff = rockVS * rockDen
292

293 #uniaxialMaterial('Viscous', matTag, C, alpha)

294 op.uniaxialMaterial('Viscous', numLayers+1, dashpotCoeff * colArea, 1)
295

296 # define dashpot element

297 op.element('zeroLength', nElemT+1, dashF, dashS, '-mat', numLayers+1, '-dir', 1)
298

299 print("Finished creating dashpot material and element...")

300

301 #-------------------------------------------------------------------------------------
˓→----
(continues on next page)

1.14. Examples 401

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

302 # 7. CREATE GRAVITY RECORDERS
303 #-------------------------------------------------------------------------------------
˓→----

304

305 # create list for pore pressure nodes

306 load_nodeList3=np.loadtxt('Node_record.txt')
307 nodeList3=[]
308

309 for i in range(len(load_nodeList3)):

310 nodeList3.append(np.int(load_nodeList3[i]))
311 # record nodal displacment, acceleration, and porepressure
312 op.recorder('Node','-file','Gdisplacement.txt','-time','-node',*nodeList3,'-dof', 1,
˓→2, 'disp')

313 op.recorder('Node','-file','Gacceleration.txt','-time','-node',*nodeList3,'-dof', 1,
˓→2, 'accel')

314 op.recorder('Node','-file','GporePressure.txt','-time','-node',*nodeList3,'-dof', 3,
˓→'vel')

315

316 # record elemental stress and strain (files are names to reflect GiD gp numbering)
317 op.recorder('Element','-file','Gstress1.txt','-time','-eleRange', 1,nElemT,'material',
˓→'1','stress')

318 op.recorder('Element','-file','Gstress2.txt','-time','-eleRange', 1,nElemT,'material',

˓→'2','stress')

319 op.recorder('Element','-file','Gstress3.txt','-time','-eleRange', 1,nElemT,'material',

˓→'3','stress')

320 op.recorder('Element','-file','Gstress4.txt','-time','-eleRange', 1,nElemT,'material',

˓→'4','stress')

321 op.recorder('Element','-file','Gstress9.txt','-time','-eleRange', 1,nElemT,'material',

˓→'9','stress')

322 op.recorder('Element','-file','Gstrain1.txt','-time','-eleRange', 1,nElemT,'material',

˓→'1','strain')

323 op.recorder('Element','-file','Gstrain2.txt','-time','-eleRange', 1,nElemT,'material',

˓→'2','strain')

324 op.recorder('Element','-file','Gstrain3.txt','-time','-eleRange', 1,nElemT,'material',

˓→'3','strain')

325 op.recorder('Element','-file','Gstrain4.txt','-time','-eleRange', 1,nElemT,'material',

˓→'4','strain')

326 op.recorder('Element','-file','Gstrain9.txt','-time','-eleRange', 1,nElemT,'material',

˓→'9','strain')

327

328 print("Finished creating gravity recorders...")

329

330 #-------------------------------------------------------------------------------------
˓→----

331 # 8. DEFINE ANALYSIS PARAMETERS

332 #-------------------------------------------------------------------------------------
˓→----

333

334 #---GROUND MOTION PARAMETERS

335 # time step in ground motion record
336 motionDT = 0.005
337 # number of steps in ground motion record
338 motionSteps = 7990
339

340 #---RAYLEIGH DAMPING PARAMETERS

341 # damping ratio
342 damp = 0.02
(continues on next page)

402 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

343 # lower frequency
344 omega1 = 2 * np.pi * 0.2
345 # upper frequency
346 omega2 = 2 * np.pi * 20
347 # damping coefficients
348 a0 = 2*damp*omega1*omega2/(omega1 + omega2)
349 a1 = 2*damp/(omega1 + omega2)
350 print("Damping Coefficients: a_0 = $a0; a_1 = $a1")
351

352 #---DETERMINE STABLE ANALYSIS TIME STEP USING CFL CONDITION

353 # maximum shear wave velocity (m/s)
354 vsMax = 250.0
355 # duration of ground motion (s)
356 duration = motionDT*motionSteps
357 # minimum element size
358 minSize = sElemY[0]
359

360 for i in range(2,numLayers+1):

361 if sElemY[i-1] <= minSize:
362 minSize = sElemY[i-1]
363

364 # trial analysis time step

365 kTrial = minSize/(vsMax**0.5)
366 # define time step and number of steps for analysis
367 if motionDT <= kTrial:
368 nSteps = motionSteps
369 dT = motionDT
370 else:
371 nSteps = np.int(mm.floor(duration/kTrial)+1)
372 dT = duration/nSteps
373

374

375 print("Number of steps in analysis: $nSteps")

376 print("Analysis time step: $dT")
377

378 #---ANALYSIS PARAMETERS

379 # Newmark parameters
380 gamma = 0.5
381 beta = 0.25
382

383 #-------------------------------------------------------------------------------------
˓→----

384 # 9. GRAVITY ANALYSIS

385 #-------------------------------------------------------------------------------------
˓→----

386 # update materials to ensure elastic behavior

387 op.updateMaterialStage('-material', 1, '-stage', 0)
388 op.updateMaterialStage('-material', 2, '-stage', 0)
389 op.updateMaterialStage('-material', 3, '-stage', 0)
390

391 op.constraints('Penalty', 1.0E14, 1.0E14)

392 op.test('NormDispIncr', 1e-4, 35, 1)
393 op.algorithm('KrylovNewton')
394 op.numberer('RCM')
395 op.system('ProfileSPD')
396 op.integrator('Newmark', gamma, beta)
397 op.analysis('Transient')
(continues on next page)

1.14. Examples 403

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

398

399 startT = tt.time()

400 op.analyze(10, 5.0E2)
401 print('Finished with elastic gravity analysis...')
402

403 # update material to consider elastoplastic behavior

404 op.updateMaterialStage('-material', 1, '-stage', 1)
405 op.updateMaterialStage('-material', 2, '-stage', 1)
406 op.updateMaterialStage('-material', 3, '-stage', 1)
407

408 # plastic gravity loading

409 op.analyze(40, 5.0e2)
410

411 print('Finished with plastic gravity analysis...')

412

413 #-------------------------------------------------------------------------------------
˓→----

414 # 10. UPDATE ELEMENT PERMEABILITY VALUES FOR POST-GRAVITY ANALYSIS

415 #-------------------------------------------------------------------------------------
˓→----

416

417 # choose base number for parameter IDs which is higer than other tags used in analysis
418 ctr = 10000.0
419 # loop over elements to define parameter IDs
420 for i in range(1,nElemT+1):
421 op.parameter(np.int(ctr+1.0), 'element', i, 'vPerm')
422 op.parameter(np.int(ctr+2.0), 'element', i, 'hPerm')
423 ctr = ctr+2.0
424

425 # update permeability parameters for each element using parameter IDs
426 ctr = 10000.0
427 for j in range(1,nElemT+1):
428 lowerBound = 0.0
429 for i in range(1,numLayers+1):
430 if j * sElemY[i-1] <= layerBound[i-1] and j*sElemY[i-1] > lowerBound:
431 op.updateParameter(np.int(ctr+1.0), vPerm[i-1])
432 op.updateParameter(np.int(ctr+2.0), hPerm[i-1])
433 lowerBound = layerBound[i-1]
434 ctr = ctr+2.0
435

436 print("Finished updating permeabilities for dynamic analysis...")

437

438 #-------------------------------------------------------------------------------------
˓→----

439 # 11. CREATE POST-GRAVITY RECORDERS

440 #-------------------------------------------------------------------------------------
˓→----

441

442 # reset time and analysis

443 op.setTime(0.0)
444 op.wipeAnalysis()
445 op.remove('recorders')
446

447 # recorder time step

448 recDT = 10*motionDT
449

450 # record nodal displacment, acceleration, and porepressure

(continues on next page)

404 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

451 op.recorder('Node','-file','displacement.txt','-time', '-dT',recDT,'-node',*nodeList3,
˓→'-dof', 1, 2, 'disp')

452 op.recorder('Node','-file','acceleration.txt','-time', '-dT',recDT,'-node',*nodeList3,

˓→'-dof', 1, 2, 'accel')

453 op.recorder('Node','-file','porePressure.txt','-time', '-dT',recDT,'-node',*nodeList3,

˓→'-dof', 3, 'vel')

454

455 # record elemental stress and strain (files are names to reflect GiD gp numbering)
456 op.recorder('Element','-file','stress1.txt','-time', '-dT',recDT,'-eleRange', 1,
˓→nElemT,'material','1','stress')

457 op.recorder('Element','-file','stress2.txt','-time', '-dT',recDT,'-eleRange', 1,

˓→nElemT,'material','2','stress')

458 op.recorder('Element','-file','stress3.txt','-time', '-dT',recDT,'-eleRange', 1,

˓→nElemT,'material','3','stress')

459 op.recorder('Element','-file','stress4.txt','-time', '-dT',recDT,'-eleRange', 1,

˓→nElemT,'material','4','stress')

460 op.recorder('Element','-file','stress9.txt','-time', '-dT',recDT,'-eleRange', 1,

˓→nElemT,'material','9','stress')

461 op.recorder('Element','-file','strain1.txt','-time', '-dT',recDT,'-eleRange', 1,

˓→nElemT,'material','1','strain')

462 op.recorder('Element','-file','strain2.txt','-time', '-dT',recDT,'-eleRange', 1,

˓→nElemT,'material','2','strain')

463 op.recorder('Element','-file','strain3.txt','-time', '-dT',recDT,'-eleRange', 1,

˓→nElemT,'material','3','strain')

464 op.recorder('Element','-file','strain4.txt','-time', '-dT',recDT,'-eleRange', 1,

˓→nElemT,'material','4','strain')

465 op.recorder('Element','-file','strain9.txt','-time', '-dT',recDT,'-eleRange', 1,

˓→nElemT,'material','9','strain')

466

467 print("Finished creating all recorders...")

468

469 #-------------------------------------------------------------------------------------
˓→----

470 # 12. DYNAMIC ANALYSIS

471 #-------------------------------------------------------------------------------------
˓→----

472 op.model('basic', '-ndm', 2, '-ndf', 3)

473

474 # define constant scaling factor for applied velocity

475 cFactor = colArea * dashpotCoeff
476

477 # define velocity time history file

478 velocityFile='velocityHistory';
479 data_gm=np.loadtxt('velocityHistory.txt')
480 #motionSteps=len(data_gm)
481 #print('Number of point for GM:',motionSteps)
482

483 # timeseries object for force history

484 op.timeSeries('Path', 2, '-dt', motionDT, '-filePath', velocityFile+'.txt', '-factor',
˓→ cFactor)

485 op.pattern('Plain', 10, 2)

486 op.load(1, 1.0, 0.0, 0.0)
487

488 print( "Dynamic loading created...")

489

490 op.constraints('Penalty', 1.0E16, 1.0E16)

491 op.test('NormDispIncr', 1e-3, 35, 1)
(continues on next page)

1.14. Examples 405

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

492 op.algorithm('KrylovNewton')
493 op.numberer('RCM')
494 op.system('ProfileSPD')
495 op.integrator('Newmark', gamma, beta)
496 op.rayleigh(a0, a1, 0.0, 0.0)
497 op.analysis('Transient')
498

499 # perform analysis with timestep reduction loop

500 ok = op.analyze(nSteps,dT)
501

502 # if analysis fails, reduce timestep and continue with analysis

503 if ok !=0:
504 print("did not converge, reducing time step")
505 curTime = op.getTime()
506 mTime = curTime
507 print("curTime: ", curTime)
508 curStep = curTime/dT
509 print("curStep: ", curStep)
510 rStep = (nSteps-curStep)*2.0
511 remStep = np.int((nSteps-curStep)*2.0)
512 print("remStep: ", remStep)
513 dT = dT/2.0
514 print("dT: ", dT)
515

516 ok = op.analyze(remStep, dT)

517 # if analysis fails again, reduce timestep and continue with analysis
518 if ok !=0:
519 print("did not converge, reducing time step")
520 curTime = op.getTime()
521 print("curTime: ", curTime)
522 curStep = (curTime-mTime)/dT
523 print("curStep: ", curStep)
524 remStep = np.int((rStep-curStep)*2.0)
525 print("remStep: ", remStep)
526 dT = dT/2.0
527 print("dT: ", dT)
528

529 ok = op.analyze(remStep, dT)

530

531 endT = tt.time()

532 print("Finished with dynamic analysis...")
533 print("Analysis execution time: ",(endT-startT))
534 op.wipe()

PM4Sand model undrained cyclic simple shear element

1. The original model is from 2D Undrained Cyclic Direct Simple Shear Test Using One Element at University
of Washington, Department of Civil and Environmental Eng by Geotechnical Eng Group L. Chen, P. Arduino -
Feb 2018.
2. The Python code is converted by Steve Xu from University of Texas at Austin ([email protected]), and
shown below, which can be downloaded here.

1 # -*- coding: utf-8 -*-

2 """
(continues on next page)

406 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

3 Created on Sat Apr 9 17:00:41 2022
4

5 @author: Zhongze Xu, The University of Texas at Austin

6

7 Openseespy code to run plane-strain stress-controlled undrained cyclic simple shear

˓→element

8 to calibrate pm4sand model

9 Basic Units are m, kN and s unless otherwise specified
10

11 original opensees code is from:

12

13 2D Undrained Cyclic Direct Simple Shear Test Using One Element

14 University of Washington, Department of Civil and Environmental Eng
15 Geotechnical Eng Group, L. Chen, P. Arduino - Feb 2018
16 Basic Units are m, kN and s unless otherwise specified
17

18 """
19 # from IPython import get_ipython;
20 # get_ipython().run_line_magic('reset', '-sf')
21

22 from datetime import datetime

23 import openseespy.opensees as op
24 import numpy as np
25 import matplotlib.pyplot as plt
26 plt.rcParams["font.family"] = "Times New Roman"
27

28 #==============================================================================
29 #Input Variables
30 '''
31 nDMaterial('PM4Sand', matTag, Dr, G0, hpo, rho, P_atm, h0, e_max, e_min,
32 nb, nd, Ado, z_max, c_z, c_e, phi_cv, nu, g_degr, c_dr, c_kaf,
33 Q, R, m_par, F_sed, p_sed)
34 '''
35 atm = -101.325
36 sig_v0 = 2.0* atm #initial vertical stress
37 CSR = 0.2 #cyclic stress ratio
38 Cycle_max = 5 #maximxum number of cycles
39 strain_in = 5.0e-6 #strain increment
40 K0 = 0.5
41 nu = K0/(1+K0) #poisson's ratio
42 devDisp = 0.03 #cutoff shear strain
43 perm = 1e-9 #permeability
44 #==============================================================================
45

46 #primary parameters
47 Dr = 0.5
48 G0 = 476.0
49 hpo = 0.53 #Contraction rate parameter
50 rho = 1.42 #mass density, KN/m3
51

52 #secondary parameters
53 P_atm = 101.325
54 # all initial stress dependant parameters have negative default values
55 # and will be calculated during initialization
56 h0 = -1.0 #Variable that adjusts the ratio of plastic modulus to elastic modulus
57 e_max = 0.8
58 e_min = 0.5
(continues on next page)

1.14. Examples 407

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

59 e_ini = e_max - (e_max - e_min)*Dr #initial void ratio
60 nb = 0.5 #Bounding surface parameter, nb>=0
61 nd = 0.1 #Dilatancy surface parameter, nd>=0
62 Ado = -1.0
63 #Dilatancy parameter, will be computed at the time of initialization if input value
˓→is negative

64 z_max = -1.0 #Fabric-dilatancy tensor parameter

65 c_z = 250.0 #Fabric-dilatancy tensor parameter
66 c_e = -1.0 #Fabric-dilatancy tensor parameter
67 phi_cv = 26.0 #Critical state effective friction angle
68 g_degr = 2.0 #Variable that adjusts degradation of elastic modulus with accumulation
˓→of fabric

69 c_dr = -1.0 #Variable that controls the rotated dilatancy surface

70 c_kaf = -1.0 # Variable that controls the effect that sustained static shear stresses
˓→have on plastic modulus

71 Q = 10.0 #Critical state line parameter

72 R = 1.5 #Critical state line parameter
73 m_par = 0.01#Yield surface constant
74 F_sed = -1.0#Variable that controls the minimum value the reduction factor of the
˓→elastic moduli can get during reconsolidation

75 p_sed = -1.0#Mean effective stress up to which reconsolidation strains are enhanced

76

77 #%%
78 #Rayleigh Damping Parameters
79 '''
80 rayleigh(alphaM, betaK, betaKinit, betaKcomm)
81 '''
82 damp = 0.02
83 omega1 = 0.2
84 omega2 = 20.0
85 a1 = 2.0*damp/(omega1+omega2) #a1 is alphaM
86 a0 = a1*omega1*omega2 #a0 is betaK
87 #%%
88 #create model
89 #Remove the existing model, important!!!
90 op.wipe()
91

92 # set modelbuilder
93 op.model('basic', '-ndm', 2, '-ndf', 3)
94

95 #model nodes
96 x1 = 0.0
97 y1 = 0.0
98

99 x2 = 1.0
100 y2 = 0.0
101

102 x3 = 1.0
103 y3 = 1.0
104

105 x4 = 0.0
106 y4 = 1.0
107

108 #create nodes

109

110 op.node(1, x1, y1)

111 op.node(2, x2, y2)
(continues on next page)

408 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

112 op.node(3, x3, y3)
113 op.node(4, x4, y4)
114

115 #boundary conditions

116 op.fix(1, 1, 1, 1)
117 op.fix(2, 1, 1, 1)
118 op.fix(3, 0, 0, 1)
119 op.fix(4, 0, 0, 1)
120 op.equalDOF(3,4,1,2) #make node 3 and 4 equal displacement at degrees 1 & 2
121

122 #material
123 #==================================================================
124 #nDMaterial('PM4Sand', matTag, D_r, G_o, h_po, Den, P_atm, h_o, e_max,
125 #e_min, n_b, n_d, A_do, z_max, c_z, c_e, phi_cv, nu, g_degr, c_dr, c_kaf,
126 #Q_bolt, R_bolt, m_par, F_sed, p_sed)
127 #==================================================================
128 op.nDMaterial('PM4Sand', 1, Dr, G0, hpo, rho, P_atm, h0, e_max, e_min,
129 nb, nd, Ado, z_max, c_z, c_e, phi_cv, nu, g_degr, c_dr, c_kaf,
130 Q, R, m_par, F_sed, p_sed)
131

132 #element
133 op.element('SSPquadUP',1, 1,2,3,4, 1, 1.0, 2.2e6, 1.0, perm, perm, e_ini, 1.0e-5)
134

135 #create recorders

136 op.recorder('Node','-file', 'Cycdisp.txt','-time', '-node',1,2,3,4,'-dof', 1, 2, 'disp
˓→')

137 op.recorder('Node','-file', 'CycPP.txt','-time', '-node',1,2,3,4,'-dof', 3, 'vel')

138 op.recorder('Element','-file', 'Cycstress.txt','-time','-ele', 1, 'stress')
139 op.recorder('Element','-file', 'Cycstrain.txt','-time','-ele', 1, 'strain')
140 #%%
141 #Analysis officially starts here
142 op.constraints('Transformation')
143 op.test('NormDispIncr', 1.0e-5, 35, 1)
144 op.algorithm('Newton')
145 op.numberer('RCM')
146 op.system('FullGeneral')
147 op.integrator('Newmark', 5.0/6.0, 4.0/9.0)
148 op.rayleigh(a1, a0, 0.0, 0.0) #modification
149 op.analysis('Transient')
150

151 #%%apply consolidation pressure

152 pNode = sig_v0/2.0 #put confining pressure evenly on two nodes
153

154 # create a plain load pattern with time series 1

155 op.timeSeries('Path', 1, '-values', 0, 1, 1, '-time', 0.0, 100.0, 1.0e10)
156 op.pattern("Plain", 1, 1, '-factor',1.0)
157 op.load(3, 0.0, pNode, 0.0) #apply vertical pressure at y direction
158 op.load(4, 0.0, pNode, 0.0)
159 op.updateMaterialStage('-material', 1, '-stage', 0)
160 op.analyze(100,1.0)
161 vDisp = op.nodeDisp(3,2)
162 b = op.eleResponse(1, 'stress') #b = [sigmaxx, sigmayy, sigmaxy]
163 print('shear stress is',b[2])
164 op.timeSeries('Path', 2, '-values', 1.0, 1.0, 1.0, '-time', 100.0, 80000.0, 1.0e10, '-
˓→factor', 1.0)

165 op.pattern('Plain', 2, 2,'-factor',1.0)

166 op.sp(3, 2, vDisp)
(continues on next page)

1.14. Examples 409

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

167 op.sp(4, 2, vDisp)
168

169 #Close Drainage

170 for i in range(4):
171 op.remove('sp', i+1, 3)
172 print('Node ID', i+1)
173

174

175 op.analyze(25,1.0)
176 b = op.eleResponse(1, 'stress') #b = [sigmaxx, sigmayy, sigmaxy]
177 print('shear stress is',b[2])
178 print('Drainage is closed')
179

180 op.updateMaterialStage('-material', 1, '-stage', 1)

181 '''
182 Note:
183 The program will use the default value of a secondary parameter if
184 a negative input is assigned to that parameter, e.g. Ado = -1.
185 However, FirstCall is mandatory when switching from elastic to elastoplastic
186 if negative inputs are assigned to stress-dependent secondary parameters,
187 e.g. Ado and zmax.
188 '''
189

190 #setParameter('-value', 0, '-ele', elementTag, 'FirstCall', matTag)

191 op.setParameter('-val', 0, '-ele', 1, 'FirstCall', '1')
192

193 op.analyze(25,1.0)
194 b = op.eleResponse(1, 'stress') #b = [sigmaxx, sigmayy, sigmaxy]
195 print('shear stress is',b[2])
196 print('finished update fixties')
197 # update Poisson's ratio for analysis
198 #setParameter -value 0.3 -ele 1 poissonRatio 1
199 op.setParameter('-val', 0.3, '-ele',1, 'poissonRatio', '1')
200

201

202 controlDisp = 1.1 * devDisp

203 numCycle = 0.25
204 print('Current Number of Cycle:', numCycle)
205

206 start = datetime.now()

207 while (numCycle <= Cycle_max):
208 #impose 1/4 cycle: zero stress to positve max stress
209 hDisp = op.nodeDisp(3,1)
210 cur_time = op.getTime()
211 steps = controlDisp/strain_in
212 time_change = cur_time + steps
213 op.timeSeries('Path', 3,'-values', hDisp, controlDisp, controlDisp, '-time', cur_
˓→time, time_change, 1.0e10, '-factor', 1.0)

214 op.pattern('Plain', 3, 3, '-fact', 1.0)

215 op.sp(3, 1, 1.0)
216 b = op.eleResponse(1, 'stress') #b = [sigmaxx, sigmayy, sigmaxy]
217 print('shear stress is',b[2])
218 while b[2] <= CSR*sig_v0*(-1.0): #b[2] is the shear stress, sigmaxy
219 op.analyze(1, 1.0)
220 b = op.eleResponse(1, 'stress')
221 hDisp = op.nodeDisp(3,1)
222 if hDisp >= devDisp:
(continues on next page)

410 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

223 print('loading break')
224 break
225 numCycle = numCycle + 0.25
226 hDisp = op.nodeDisp(3,1)
227 cur_time = op.getTime()
228 op.remove('loadPattern', 3)
229 op.remove('timeSeries', 3)
230 op.remove('sp', 3, 1)
231 #impose 1/2 cycle: Postive max stress to negative max stress
232 steps = (controlDisp+hDisp)/strain_in
233 time_change = cur_time + steps
234 op.timeSeries('Path', 3,'-values', hDisp, -1.0*controlDisp, -1.0*controlDisp, '-
˓→time', cur_time, time_change, 1.0e10, '-factor', 1.0)

235 op.pattern('Plain', 3, 3)
236 op.sp(3, 1, 1.0)
237 while b[2] > CSR*sig_v0:
238 op.analyze(1, 1.0)
239 b = op.eleResponse(1, 'stress')
240 print('shear stress is',b[2])
241 hDisp = op.nodeDisp(3,1)
242 if hDisp <= -1.0*devDisp:
243 print('unloading break')
244 break
245 numCycle = numCycle + 0.5
246 hDisp = op.nodeDisp(3,1)
247 cur_time = op.getTime()
248 op.remove('loadPattern', 3)
249 op.remove('timeSeries', 3)
250 op.remove('sp', 3, 1)
251 #impose 1/4 cycle
252 steps = (controlDisp+hDisp)/strain_in
253 op.timeSeries('Path', 3,'-values', hDisp, controlDisp, controlDisp, '-time', cur_
˓→time, time_change, 1.0e10, '-factor', 1.0)

254 op.pattern('Plain', 3, 3, '-fact', 1.0)

255 op.sp(3, 1, 1.0)
256 while b[2] <= 0.0: #b[2] is the shear stress, sigmaxy
257 op.analyze(1, 1.0)
258 b = op.eleResponse(1, 'stress')
259 print('shear stress is',b[2])
260 hDisp = op.nodeDisp(3,1)
261 if hDisp >= devDisp:
262 print('reloading break')
263 break
264 numCycle = numCycle + 0.25
265 print('Current Number of Cycle:', numCycle)
266 op.remove('loadPattern', 3)
267 op.remove('timeSeries', 3)
268 #op.remove('sp', 3, 1)
269

270 op.wipe()
271 print('Analysis is done!')
272 end = datetime.now()
273 run_time = end-start
274 print('Computation time is' , run_time)
275 #%%PostProcessing
276 import pandas as pd
277 df_stress = pd.read_csv('Cycstress.txt', sep=" ", header=None)
(continues on next page)

1.14. Examples 411

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

278 df_strain = pd.read_csv('Cycstrain.txt', sep=" ", header=None)
279 #df_PP = pd.read_csv('CycPP.txt', sep=" ", header=None)
280

281 Stress_V = df_stress.iloc[:, 1].to_numpy()*(-1.0) #compression is positive

282 Shear_Stress = df_stress.iloc[:, 3].to_numpy()
283 Shear_Strain = df_strain.iloc[:, 3].to_numpy()*100.0
284

285 fig = plt.figure(figsize=(10,18))

286 ax0 = fig.add_subplot(211)
287 ax0.plot(Shear_Strain, Shear_Stress, label='stress-strain', linewidth=0.8)
288 ax0.set_xlabel("Shear Strain,%", fontsize=16)
289 ax0.set_ylabel("Shear Stress, kPa", fontsize=16)
290 ax0.legend(fontsize=16)
291

292 ax1 = fig.add_subplot(212)

293 ax1.plot(Stress_V, Shear_Stress, label='Stress Path', linewidth=0.8)
294 ax1.set_xlabel("Vertical Stress, kPa", fontsize=16)
295 ax1.set_ylabel("Shear Stress, kPa", fontsize=16)
296 ax1.legend(fontsize=16)
297 plt.show()
298 plt.close()

1.14.5 Thermal Examples

1. Restrained beam under thermal expansion

Restrained beam under thermal expansion

1. The original model can be found here.

2. The Pypton source code is shown below, which can be downloaded here.
3. Make sure the numpy and matplotlib packages are installed in your Python distribution.
4. Run the source code in your favorate Python program and should see

412 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1 from openseespy.opensees import *

2

3 import numpy as np
4 import matplotlib.pyplot as plt
5

6 # define model
7 model('basic', '-ndm', 2, '-ndf', 3)
8

9 #define node
10 node(1, 0.0, 0.0)
11 node(2, 2.0, 0.0)
12 node(3, 1.0, 0.0)
13

14 #define boundary condition

15 fix(1, 1, 1, 1)
16 fix(2, 1, 1, 1)
17 fix(3, 0, 1, 1)
18

19 #define an elastic material with Tag=1 and E=2e11.

20 matTag = 1
21 uniaxialMaterial('Steel01Thermal', 1, 2e11, 2e11, 0.01)
22

23 #define fibred section Two fibres: fiber $yLoc $zLoc $A $matTag

24 secTag = 1
25 section('FiberThermal',secTag)
(continues on next page)

1.14. Examples 413

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

26 fiber(-0.025, 0.0, 0.005, matTag)
27 fiber(0.025, 0.0, 0.005, matTag)
28

29 #define coordinate transforamtion

30 #three transformation types can be chosen: Linear, PDelta, Corotational)
31 transfTag = 1
32 geomTransf('Linear', transfTag)
33

34 # beam integration
35 np = 3
36 biTag = 1
37 beamIntegration('Lobatto',biTag, secTag, np)
38

39 #define beam element

40 element('dispBeamColumnThermal', 1, 1, 3, transfTag, biTag)
41 element('dispBeamColumnThermal', 2, 3, 2, transfTag, biTag)
42

43 # define time series

44 tsTag = 1
45 timeSeries('Linear',tsTag)
46

47 # define load pattern

48 patternTag = 1
49 maxtemp = 1000.0
50 pattern('Plain', patternTag, tsTag)
51 eleLoad('-ele', 1, '-type', '-beamThermal', 1000.0, -0.05, 1000.0, 0.05)
52 #eleLoad -ele 2 -type -beamThermal 0 -0.05 0 0.05
53

54 # define analysis
55 incrtemp = 0.01
56 system('BandGeneral')
57 constraints('Plain')
58 numberer('Plain')
59 test('NormDispIncr', 1.0e-3, 100, 1)
60 algorithm('Newton')
61 integrator('LoadControl', incrtemp)
62 analysis('Static')
63

64 # analysis
65 nstep = 100
66 temp = [0.0]
67 disp = [0.0]
68 for i in range(nstep):
69 if analyze(1) < 0:
70 break
71

72 temp.append(getLoadFactor(patternTag)*maxtemp)
73 disp.append(nodeDisp(3,1))
74

75

76 plt.plot(temp,disp,'-o')
77 plt.xlabel('Temperature')
78 plt.ylabel('Nodal displacement')
79 plt.grid()
80 plt.show()

414 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1.14.6 Parallel Examples

1. Hello World Example 1

2. Hello World Example 2
3. Parallel Truss Example
4. Parallel Tri31 Example
5. Parallel Parametric Study Example

Hello World Example 1

1. The source code is shown below, which can be downloaded here.

2. Run the source code with 4 processors

mpiexec -np 4 python hello.py

the outputs look like

Hello World Process: 1

Hello World Process: 2
Hello World Process: 0
Total number of processes: 4
Hello World Process: 3
Process 1 Terminating
Process 2 Terminating
Process 0 Terminating
Process 3 Terminating

The script is shown below

1 import openseespy.opensees as ops

2

3 pid = ops.getPID()
4 np = ops.getNP()
5

6 print('Hello World Process:', pid)

7 if pid == 0:
8 print('Total number of processes:', np)
9

Hello World Example 2

1. The source code is shown below, which can be downloaded here.

2. Run the source code with 4 processors

mpiexec -np 4 python hello2.py

the outputs look like

Random:
Hello from 2
Hello from 1
(continues on next page)

1.14. Examples 415

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

Hello from 3

Ordered:
Hello from 1
Hello from 2
Hello from 3

Broadcasting:
Hello from 0
Hello from 0
Hello from 0
Process 3 Terminating
Process 2 Terminating
Process 1 Terminating
Process 0 Terminating

The script is shown below

1 import openseespy.opensees as ops
2

3 pid = ops.getPID()
4 np = ops.getNP()
5

6 # datatype = 'float'
7 # datatype = 'int'
8 datatype = 'str'
9

10 if pid == 0:
11 print('Random: ')
12

13 for i in range(1, np):

14 data = ops.recv('-pid', 'ANY')
15 print(data)
16 else:
17 if datatype == 'str':
18 ops.send('-pid', 0, 'Hello from {}'.format(pid))
19 elif datatype == 'float':
20 ops.send('-pid', 0, float(pid))
21 elif datatype == 'int':
22 ops.send('-pid', 0, int(pid))
23

24 ops.barrier()
25

26 if pid == 0:
27 print('\nOrdered: ')
28

29 for i in range(1, np):

30 data = ops.recv('-pid', i)
31 print(data)
32 else:
33 if datatype == 'str':
34 ops.send('-pid', 0, 'Hello from {}'.format(pid))
35 elif datatype == 'float':
36 ops.send('-pid', 0, float(pid))
37 elif datatype == 'int':
38 ops.send('-pid', 0, int(pid))
39
(continues on next page)

416 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

40 ops.barrier()
41 if pid == 0:
42 print('\nBroadcasting: ')
43 if datatype == 'str':
44 ops.Bcast('Hello from {}'.format(pid))
45 elif datatype == 'float':
46 ops.Bcast(float(pid))
47 elif datatype == 'int':
48 ops.Bcast(int(pid))
49 else:
50 data = ops.Bcast()
51 print(data)

Parallel Truss Example

1. The source code is shown below, which can be downloaded here.

2. Run the source code with 2 processors

mpiexec -np 2 python paralleltruss.py

the outputs look like

Node 4: [[72.0, 96.0], [0.5300927771322836, -0.17789363846931772]]

Node 4: [[72.0, 96.0], [0.5300927771322836, -0.17789363846931772]]
Node 4: [[72.0, 96.0], [1.530092777132284, -0.19400676316761836]]
Node 4: [[72.0, 96.0], [1.530092777132284, -0.19400676316761836]]
opensees.msg: TIME(sec) Real: 0.208238

opensees.msg: TIME(sec) Real: 0.209045

Process 0 Terminating
Process 1 Terminating

The script is shown below

1 import openseespy.opensees as ops

2

3 pid = ops.getPID()
4 np = ops.getNP()
5 ops.start()
6 if np != 2:
7 exit()
8

9 ops.model('basic', '-ndm', 2, '-ndf', 2)

10 ops.uniaxialMaterial('Elastic', 1, 3000.0)
11

12 if pid == 0:
13 ops.node(1, 0.0, 0.0)
14 ops.node(4, 72.0, 96.0)
15

16 ops.fix(1, 1, 1)
17

18 ops.element('Truss', 1, 1, 4, 10.0, 1)
19 ops.timeSeries('Linear', 1)
(continues on next page)

1.14. Examples 417

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

20 ops.pattern('Plain', 1, 1)
21 ops.load(4, 100.0, -50.0)
22

23 else:
24 ops.node(2, 144.0, 0.0)
25 ops.node(3, 168.0, 0.0)
26 ops.node(4, 72.0, 96.0)
27

28 ops.fix(2, 1, 1)
29 ops.fix(3, 1, 1)
30

31 ops.element('Truss', 2, 2, 4, 5.0, 1)
32 ops.element('Truss', 3, 3, 4, 5.0, 1)
33

34 ops.constraints('Transformation')
35 ops.numberer('ParallelPlain')
36 ops.system('Mumps')
37 ops.test('NormDispIncr', 1e-6, 6, 2)
38 ops.algorithm('Newton')
39 ops.integrator('LoadControl', 0.1)
40 ops.analysis('Static')
41

42 ops.analyze(10)
43

44 print('Node 4: ', [ops.nodeCoord(4), ops.nodeDisp(4)])

45

46 ops.loadConst('-time', 0.0)
47

48 if pid == 0:
49 ops.pattern('Plain', 2, 1)
50 ops.load(4, 1.0, 0.0)
51

52 ops.domainChange()
53 ops.integrator('ParallelDisplacementControl', 4, 1, 0.1)
54 ops.analyze(10)
55

56 print('Node 4: ', [ops.nodeCoord(4), ops.nodeDisp(4)])

57 ops.stop()

Parallel Tri31 Example

1. The source code is shown below, which can be downloaded here.

2. Run the source code with 4 processors

mpiexec -np 4 python paralleltri31.py

the outputs look like

opensees.msg: TIME(sec) Real: 0.177647

opensees.msg: TIME(sec) Real: 0.187682

opensees.msg: TIME(sec) Real: 0.193193

(continues on next page)

418 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

opensees.msg: TIME(sec) Real: 0.19473

opensees.msg: TIME(sec) Real: 14.4652

Node 4 [-0.16838893553441528, -2.88399389660282]

opensees.msg: TIME(sec) Real: 14.4618

opensees.msg: TIME(sec) Real: 14.4619

opensees.msg: TIME(sec) Real: 14.4948

Process 0 Terminating
Process 1 Terminating
Process 2 Terminating
Process 3 Terminating

The script is shown below

1 import openseespy.opensees as ops
2

3 pid = ops.getPID()
4 np = ops.getNP()
5 ops.start()
6

7 ops.model('basic', '-ndm', 2, '-ndf', 2)

8

9 L = 48.0
10 H = 4.0
11

12 Lp = L / np
13 ndf = 2
14 meshsize = 0.05
15

16 ops.node(pid, Lp * pid, 0.0)

17 ops.node(pid + 1, Lp * (pid + 1), 0.0)
18 ops.node(np + pid + 2, Lp * (pid + 1), H)
19 ops.node(np + pid + 1, Lp * pid, H)
20

21 sid = 1
22 ops.setStartNodeTag(2 * np + 2 + pid * int(H / meshsize + 10))
23 ops.mesh('line', 3, 2, pid, np + pid + 1, sid, ndf, meshsize)
24 ops.setStartNodeTag(2 * np + 2 + (pid + 1) * int(H / meshsize + 10))
25 ops.mesh('line', 4, 2, pid + 1, np + pid + 2, sid, ndf, meshsize)
26

27 ops.setStartNodeTag(int(2 * L / meshsize + (np + 1) * H / meshsize * 2) +

28 pid * int(H * L / meshsize ** 2 * 2))
29 ops.mesh('line', 1, 2, pid, pid + 1, sid, ndf, meshsize)
30 ops.mesh('line', 2, 2, np + pid + 1, np + pid + 2, sid, ndf, meshsize)
31

32 ops.nDMaterial('ElasticIsotropic', 1, 3000.0, 0.3)

33

34 eleArgs = ['tri31', 1.0, 'PlaneStress', 1]

35

36 ops.mesh('quad', 5, 4, 1, 4, 2, 3, sid, ndf, meshsize, *eleArgs)

37

38
(continues on next page)

1.14. Examples 419

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

39 if pid == 0:
40 ops.fix(pid, 1, 1)
41 ops.fix(np+pid+1, 1, 1)
42 if pid == np-1:
43 ops.timeSeries('Linear', 1)
44 ops.pattern('Plain', 1, 1)
45 ops.load(np + pid + 2, 0.0, -1.0)
46

47

48 ops.constraints('Transformation')
49 ops.numberer('ParallelPlain')
50 ops.system('Mumps')
51 ops.test('NormDispIncr', 1e-6, 6)
52 ops.algorithm('Newton')
53 ops.integrator('LoadControl', 1.0)
54 ops.analysis('Static')
55

56 ops.stop()
57 ops.start()
58 ops.analyze(1)
59

60 if pid == np-1:
61 print('Node', pid+1, ops.nodeDisp(pid+1))
62

63

64 ops.stop()

Parallel Parametric Study Example

1. The source code is shown below, which can be downloaded here.

2. Run the source code with 2 processors

mpiexec -np 2 python paralleltruss2.py

the outputs look like

Processor 0
Node 4 (E = 3000.0 ) Disp : [0.5300927771322836, -0.17789363846931766]
Processor 1

Node 4 (E = 6000.0 ) Disp : [0.2650463885661418, -0.08894681923465883]

Process 1 Terminating
Process 0 Terminating

The script is shown below

1 import openseespy.opensees as ops

2

3 pid = ops.getPID()
4 np = ops.getNP()
5 ops.start()
6 if np != 2:
(continues on next page)

420 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

7 exit()
8

9 ops.model('basic', '-ndm', 2, '-ndf', 2)

10

11 if pid == 0:
12 E = 3000.0
13 else:
14 E = 6000.0
15

16 ops.uniaxialMaterial('Elastic', 1, E)
17

18 ops.node(1, 0.0, 0.0)

19 ops.node(2, 144.0, 0.0)
20 ops.node(3, 168.0, 0.0)
21 ops.node(4, 72.0, 96.0)
22

23 ops.fix(1, 1, 1)
24 ops.fix(2, 1, 1)
25 ops.fix(3, 1, 1)
26

27 ops.element('Truss', 1, 1, 4, 10.0, 1)
28 ops.timeSeries('Linear', 1)
29 ops.pattern('Plain', 1, 1)
30 ops.load(4, 100.0, -50.0)
31

32 ops.element('Truss', 2, 2, 4, 5.0, 1)
33 ops.element('Truss', 3, 3, 4, 5.0, 1)
34

35 ops.constraints('Transformation')
36 ops.numberer('ParallelPlain')
37 ops.system('Umfpack')
38 ops.test('NormDispIncr', 1e-6, 6)
39 ops.algorithm('Newton')
40 ops.integrator('LoadControl', 0.1)
41 ops.analysis('Static')
42

43 ops.analyze(10)
44

45

46 if pid == 0:
47 print('Processor 0')
48 print('Node 4 (E =', E, ') Disp :', ops.nodeDisp(4))
49

50 ops.barrier()
51

52 if pid == 1:
53 print('Processor 1')
54 print('Node 4 (E =', E, ') Disp :', ops.nodeDisp(4))
55

56

57 ops.stop()

1.14.7 Plotting Examples

1. vfo - Visualization for OpenSees (Examples)

1.14. Examples 421

OpenSeesPy Documentation, Release 3.5.1.3

2. opsvis Examples

vfo - Visualization for OpenSees (Examples)

Note: Visit the documentation page of vfo to see examples of OpenSees model post-processing.

opsvis Examples

Note: The opsvis module has been moved to its own respository. The examples can be found there.

1.14.8 Sensitivity Examples

1. Simple cantilever for load-controlled sensitivity analysis

2. Two storey steel moment frame with W-sections for displacement-controlled sensitivity analysis

Simple cantilever for load-controlled sensitivity analysis

1. The source code is developed by Marin Grubišić at University of Osijek, Croatia.

2. The numerical model with the associated analysis was described in detail by Prof. Michael Scott within
OpenSees Days 2011.
3. The source code is shown below, which can be downloaded here.
4. Run the source code in your favorite Python program and should see following plot.

422 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1.14. Examples 423

 OpenSeesPy Documentation, Release 3.5.1.3

1 """
2 - The source code is developed by Marin Grubišić https://github.com/mgrubisic
3 at University of Osijek, Croatia.
4 - The numerical model with the associated analysis was described in detail by
5 Prof. Michael Scott within OpenSees Days 2011
6 https://opensees.berkeley.edu/OpenSees/workshops/OpenSeesDays2011/B5_MHS.pdf
7 - Run the source code in your favorite Python program and should see following plot.
8 """
9

10 import time
11 import sys
12 import numpy as np
13 import matplotlib.pyplot as plt
14 import openseespy.opensees as ops
15

16 # +===============================================================================+
17 # | OpenSees Header |
18 # +===============================================================================+
19 nSpaces = 90
20 OpenSeesHeader = {"header_00": " ",
21 "header_01": nSpaces * "=",
22 "header_02": "OpenSees -- Open System For Earthquake Engineering
˓→Simulation",

23 "header_03": "Pacific Earthquake Engineering Research Center (PEER)

˓→",

24 "header_04": "OpenSees " + ops.version() + " 64-Bit",

25 "header_05": "Python " + sys.version,
26 "header_06": " ",
27 "header_07": "(c) Copyright 1999-2021 The Regents of the University
˓→of California",

28 "header_08": "All Rights Reserved",

29 "header_09": "(Copyright and Disclaimer @ http://www.berkeley.edu/
˓→OpenSees/copyright.html)",

30 "header_10": nSpaces * "=",

31 "header_11": " ",
32 }
33 for i in OpenSeesHeader.keys():
34 print(OpenSeesHeader[i].center(nSpaces, " "))
35

36

37 def title(title="Title Example", nSpaces=nSpaces):

38 header = (nSpaces-2) * "-"
39 print("+" + header.center((nSpaces-2), " ") + "+")
40 print("|" + title.center((nSpaces-2), " ") + "|")
41 print("+" + header.center((nSpaces-2), " ") + "+")
42

43

44 # +===============================================================================+
45 # | Units |
46 # +===============================================================================+
47 m, kN, sec = 1.0, 1.0, 1.0 # meter for length, kilonewton for force, second for time
48

49 # Angle
50 rad = 1.0
51 deg = np.pi/180.0*rad
52

53 # Length, Area, Volume, Second moment of area

(continues on next page)

424 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

54 m2, m3, m4 = m**2, m**3, m**4
55 cm, cm2, cm3, cm4 = m*1E-2, m*1E-4, m*1E-6, m*1E-8
56 mm, mm2, mm3, mm4 = m*1E-3, m*1E-6, m*1E-9, m*1E-12
57 inch = 0.0254*m
58 ft = 0.3048*m
59

60 # Force
61 N = kN*1E-3
62 g = 9.80665*m/(sec**2)
63

64 # Mass
65 kg = N*sec**2/m
66 ton = kg*1E3
67 lbs = 0.45359237*kg
68 kip = 453.59237*kg
69

70 # Pressure
71 Pa, kPa, MPa, GPa = N/m**2, 1E3*N/m**2, 1E6*N/m**2, 1E9*N/m**2
72 pcf = lbs/(ft**3)
73 ksi = kip/(inch**2)
74 psi = ksi/1E3
75

76 Inf = 1.0E12 # a really large number

77 Null = 1/Inf # a really small number
78

79 LunitTXT = "m" # (Length) define basic-unit text for output

80 FunitTXT = "kN" # (Force) define basic-unit text for output
81 TunitTXT = "seconds" # (Time) define basic-unit text for output
82

83 # +===============================================================================+
84 # | Define some functions |
85 # +===============================================================================+
86

87

88 def run_sensitivity_analysis(ctrlNode, dof, baseNode, SensParam, steps=500,

˓→IOflag=False):

89 """
90 Run load-control sensitivity analysis
91 """
92 ops.wipeAnalysis()
93 start_time = time.time()
94

95 title("Running Load-Control Sensitivity Analysis ...")

96

97 ops.system("BandGeneral")
98 ops.numberer("RCM")
99 ops.constraints("Transformation")
100 ops.test("NormDispIncr", 1.0E-12, 10, 3)
101 ops.algorithm("Newton") # KrylovNewton
102 ops.integrator("LoadControl", 1/steps)
103 ops.analysis("Static")
104 ops.sensitivityAlgorithm("-computeAtEachStep") # automatically compute
˓→sensitivity at the end of each step

105

106 outputs = {"time": np.array([]),

107 "disp": np.array([]),
108 "force": np.array([]),
(continues on next page)

1.14. Examples 425

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

109 }
110

111 for sens in SensParam:

112 outputs[f"sensDisp_{sens}"] = np.array([]),
113

114 for i in range(steps):

115 ops.reactions()
116 if IOflag:
117 print(
118 f"Single Cycle Response: Step #{i}, Node #{ctrlNode}: {ops.
˓→nodeDisp(ctrlNode, dof):.3f} {LunitTXT} / {-ops.nodeReaction(baseNode, dof):.2f}

˓→{FunitTXT}.")

119 ops.analyze(1)
120 tCurrent = ops.getTime()
121

122 outputs["time"] = np.append(outputs["time"], tCurrent)

123 outputs["disp"] = np.append(outputs["disp"], ops.nodeDisp(ctrlNode, dof))
124 outputs["force"] = np.append(outputs["force"], -ops.nodeReaction(baseNode,
˓→ dof))
125

126 for sens in SensParam:

127 # sensDisp(patternTag, paramTag)
128 outputs[f"sensDisp_{sens}"] = np.append(outputs[f"sensDisp_{sens}"], ops.
˓→sensNodeDisp(ctrlNode, dof, sens))

129

130 title("Sensitvity Analysis Completed!")

131 print(f"Analysis elapsed time is {(time.time() - start_time):.3f} seconds.\n")
132

133 return outputs

134

135

136 # +===============================================================================+
137 # | Define model |
138 # +===============================================================================+
139 # Create ModelBuilder
140 # -------------------
141 ops.wipe()
142 ops.model("basic", "-ndm", 2, "-ndf", 3)
143

144 # Create nodes

145 # ------------
146 L = 5*m
147 ops.node(1, 0.0, 0.0) # Fixed end
148 ops.node(2, L, 0.0) # Free end
149

150 # Fixed support

151 # -------------
152 ops.fix(1, 1, 1, 1)
153

154 # Define material

155 # ---------------
156 matTag = 1
157 Fy = 410*MPa # Yield stress
158 Es = 200*GPa # Modulus of Elasticity of Steel
159 b = 2/100 # 2% Strain hardening ratio
160 Hkin = b/(1-b)*Es
161
(continues on next page)

426 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

162 # Sensitivity-ready steel materials: Hardening, Steel01, SteelMP, BoucWen, SteelBRB,
˓→StainlessECThermal, SteelECThermal, ...

163 # Hardening Sensitivity Params: sigmaY/fy/Fy, E, H_kin/Hkin, H_iso/Hiso

164 ops.uniaxialMaterial("Hardening", matTag, Es, Fy, 0, Hkin)
165

166 # ops.uniaxialMaterial("Steel01", matTag, Fy, Es, b) # Sensitivity Params: sigmaY/fy/

˓→Fy, E, b, a1, a2, a3, a4

167 # ops.uniaxialMaterial("SteelMP", matTag, Fy, Es, b) # Sensitivity Params: sigmaY/fy,

˓→E, b

168

169 # Define sections

170 # ---------------
171 # Sections defined with "canned" section ("WFSection2d"), otherwise use a
˓→FiberSection object (ops.section("Fiber",...))

172 beamSecTag = 1
173 beamWidth, beamDepth = 10*cm, 50*cm
174 # secTag, matTag, d, tw, bf, tf,
˓→Nfw, Nff

175 ops.section("WFSection2d", beamSecTag, matTag, beamDepth, beamWidth, beamWidth, 0, 20,

˓→ 0) # Beam section
176

177 # Define elements

178 # ---------------
179 beamTransTag, beamIntTag = 1, 1
180 # Linear, PDelta, Corotational
181 ops.geomTransf("Corotational", beamTransTag)
182

183 nip = 5
184 # Lobatto, Legendre, NewtonCotes, Radau, Trapezoidal, CompositeSimpson
185 ops.beamIntegration("Legendre", beamIntTag, beamSecTag, nip)
186

187 # Beam elements

188 numEle = 5
189 meshSize = L/numEle # mesh size
190

191 eleType = "dispBeamColumn" # forceBeamColumn, "dispBeamColumn"

192 # tag, Npts, nodes, type, dofs, size, eleType, transfTag, beamIntTag
193 ops.mesh("line", 1, 2, *[1, 2], 0, 3, meshSize, eleType, beamTransTag, beamIntTag)
194

195 # Create a Plain load pattern with a Sine/Trig TimeSeries

196 # -------------------------------------------------------
197 # tag, tStart, tEnd, period, factor
198 ops.timeSeries("Trig", 1, 0.0, 1.0, 1.0, "-factor", 1.0) # "Sine", "Trig" or
˓→"Triangle"

199 ops.pattern("Plain", 1, 1)
200

201 P = 1710*kN
202 # Create nodal loads at node 2
203 # nd FX FY MZ
204 ops.load(2, 0.0, P, 0.0)
205

206 # +===============================================================================+
207 # | Define Sensitivity Parameters |
208 # +===============================================================================+
209 # /// Each parameter must be unique in the FE domain, and all parameter tags MUST be
˓→numbered sequentially starting from 1! ///

210 ops.parameter(1) # Blank parameters

(continues on next page)

1.14. Examples 427

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

211 ops.parameter(2)
212 ops.parameter(3)
213 ops.parameter(4)
214 for ele in range(1, numEle+1): # Only column elements
215 ops.addToParameter(1, "element", ele, "E") # E
216 # Check the sensitivity parameter names in *.cpp files ("sigmaY" or "fy" or "Fy")
217 # https://github.com/OpenSees/OpenSees/blob/master/SRC/material/uniaxial/
˓→HardeningMaterial.cpp

218 ops.addToParameter(2, "element", ele, "Fy") # "sigmaY" or "fy" or "Fy"

219 ops.addToParameter(3, "element", ele, "Hkin") # "H_kin" or "Hkin" or "b"
220 ops.addToParameter(4, "element", ele, "d") # "d"
221

222 ops.parameter(5, "node", 2, "coord", 1) # parameter for coordinate of node 2 in DOF

˓→"1" (PX=1, PY=2, MZ=3)

223 # Map parameter 6 to vertical load at node 2 contained in load pattern 1 (last
˓→argument is global DOF, e.g., in 2D PX=1, PY=2, MZ=3)

224 ops.parameter(6, "loadPattern", 1, "loadAtNode", 2, 2)

225

226 ParamSym = ["E", "F_y", "H_{kin}", "d", "L", "P"]

227 ParamVars = [Es, Fy, Hkin, beamDepth, L, P]
228

229 title("Model Built")

230

231 # +===============================================================================+
232 # | Run the analysis |
233 # +===============================================================================+
234 # Run analysis with 500 steps
235 # -------------------------
236 outputs = run_sensitivity_analysis(
237 ctrlNode=2, dof=2, baseNode=1, SensParam=ops.getParamTags(), steps=500,
˓→IOflag=False)

238

239 # +===============================================================================+
240 # | Plot results |
241 # +===============================================================================+
242 rows, columns = 7, 2
243 grid = plt.GridSpec(rows, columns, wspace=0.25, hspace=0.25)
244 plt.figure(figsize=(10, 15))
245

246

247 def plot_params():

248 plt.rc("axes", axisbelow=True)
249 plt.tick_params(direction="in", length=5, colors="k", width=0.75)
250 plt.grid(True, color="silver", linestyle="solid",
251 linewidth=0.75, alpha=0.75)
252

253

254 # Subplot #1
255 # ----------
256 plt.subplot(grid[0]), plot_params()
257 plt.plot(outputs["time"],
258 outputs["force"], "-k", linewidth=1.5)
259 plt.ylabel(r"Load, $P$ [kN]")
260

261 # Subplot #2
262 # ----------
263 plt.subplot(grid[1]), plot_params()
(continues on next page)

428 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

264 plt.plot(outputs["disp"],
265 outputs["force"], "-k", linewidth=2.0, label="$U$")
266 plt.ylabel(r"Load, $P$ [kN]"), plt.legend(fontsize=9)
267

268 i, j = 2, 0
269 for p in ParamVars:
270 # Subplot #i
271 # ----------
272 plt.subplot(grid[i]), plot_params()
273 plt.plot(outputs["time"], outputs[f"sensDisp_{j+1}"]*p, "-.k", linewidth=1.5,
˓→label="DDM")

274 plt.fill_between(outputs["time"], outputs[f"sensDisp_{j+1}"]*p, color='grey',

˓→alpha=0.15)

275 plt.ylabel(f"$(\partial U/\partial {ParamSym[j]}){ParamSym[j]}$ [m]")

276 plt.legend(fontsize=9)
277 if j == 5:
278 plt.xlabel(r"Time, $t$")
279

280 # Subplot #ii

281 # -----------
282 plt.subplot(grid[i+1]), plot_params()
283 plt.plot(outputs["disp"], outputs["force"], "-k", linewidth=2.0, label="$U$")
284 plt.plot(outputs["disp"] + outputs[f"sensDisp_{j+1}"] * 0.1*p,
285 outputs["force"], "--k", linewidth=1.5, label=f"$U + (\partial U/
˓→\partial {ParamSym[j]})0.1{ParamSym[j]}$")

286 plt.plot(outputs["disp"] - outputs[f"sensDisp_{j+1}"] * 0.1*p,

287 outputs["force"], "-.k", linewidth=1.5, label=f"$U - (\partial U/
˓→\partial {ParamSym[j]})0.1{ParamSym[j]}$")

288

289 plt.fill_betweenx(outputs["force"], outputs["disp"] +

290 outputs[f"sensDisp_{j+1}"] * 0.1*p, outputs["disp"] - outputs[f
˓→"sensDisp_{j+1}"] * 0.1*p, color='grey', alpha=0.15)

291

292 plt.ylabel(r"Load, $P$ [kN]")

293 plt.legend(fontsize=9)
294 if j == 5:
295 plt.xlabel(r"Displacement, $U$ [m]")
296 i += 2
297 j += 1
298

299

300 # Save figure

301 # -----------
302 plt.savefig("CantileverSensitivity2D_v1.png", bbox_inches="tight", pad_inches=0.05,
˓→dpi=300, format="png")

303 plt.show()

Two storey steel moment frame with W-sections for displacement-controlled sensitivity analysis

1. The source code is developed by Marin Grubišić at University of Osijek, Croatia.

2. The numerical model with the associated analysis was described in detail by Prof. Michael Scott in the Portwood
Digital blog.
3. The source code is shown below, which can be downloaded here.
4. Run the source code in your favorite Python program and should see following plot.

1.14. Examples 429

OpenSeesPy Documentation, Release 3.5.1.3

430 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

1 """
2 - The source code is developed by Marin Grubišić https://github.com/mgrubisic
3 at University of Osijek, Croatia.
4 - The numerical model with the associated analysis was described in detail by
5 Prof. Michael Scott in the Portwood Digital blog
6 https://portwooddigital.com/2021/01/03/sensitivity-training/
7 - Run the source code in your favorite Python program and should see following plot.
8 """
9

10 import time
11 import sys
12 import numpy as np
13 import matplotlib.pyplot as plt
14 import openseespy.opensees as ops
15

(continues on next page)

1.14. Examples 431

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

16 # +===============================================================================+
17 # | OpenSees Header |
18 # +===============================================================================+
19 nSpaces = 90
20 OpenSeesHeader = {"header_00": " ",
21 "header_01": nSpaces * "=",
22 "header_02": "OpenSees -- Open System For Earthquake Engineering
˓→Simulation",

23 "header_03": "Pacific Earthquake Engineering Research Center (PEER)

˓→",

24 "header_04": "OpenSees " + ops.version() + " 64-Bit",

25 "header_05": "Python " + sys.version,
26 "header_06": " ",
27 "header_07": "(c) Copyright 1999-2021 The Regents of the University
˓→of California",

28 "header_08": "All Rights Reserved",

29 "header_09": "(Copyright and Disclaimer @ http://www.berkeley.edu/
˓→OpenSees/copyright.html)",

30 "header_10": nSpaces * "=",

31 "header_11": " ",
32 }
33 for i in OpenSeesHeader.keys():
34 print(OpenSeesHeader[i].center(nSpaces, " "))
35

36

37 def title(title="Title Example", nSpaces=nSpaces):

38 header = (nSpaces-2) * "-"
39 print("+" + header.center((nSpaces-2), " ") + "+")
40 print("|" + title.center((nSpaces-2), " ") + "|")
41 print("+" + header.center((nSpaces-2), " ") + "+")
42

43

44 # +===============================================================================+
45 # | Units |
46 # +===============================================================================+
47 inch = 1.0 # define basic units
48 kip = 1.0
49 sec = 1.0
50

51 ft = 12*inch
52 lb = kip/1000
53 ksi = kip/inch**2
54 psf = lb/ft**2
55

56 LunitTXT = "inches" # (Length) define basic-unit text for output

57 FunitTXT = "kips" # (Force) define basic-unit text for output
58 TunitTXT = "seconds" # (Time) define basic-unit text for output
59

60 # +===============================================================================+
61 # | Define some functions |
62 # +===============================================================================+
63

64

65 def run_gravity_analysis(steps=10):
66 """
67 Run gravity analysis (in 10 steps)
68 """
(continues on next page)

432 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

69 ops.wipeAnalysis()
70 ops.system("BandGeneral")
71 ops.numberer("RCM")
72 ops.constraints("Transformation")
73 ops.test("NormDispIncr", 1.0E-12, 10, 3)
74 ops.algorithm("Newton") # KrylovNewton
75 ops.integrator("LoadControl", 1/steps)
76 ops.analysis("Static")
77 ops.analyze(steps)
78 title("Gravity Analysis Completed!")
79 # Set the gravity loads to be constant & reset the time in the domain
80 ops.loadConst("-time", 0.0)
81 ops.wipeAnalysis()
82

83

84 def run_sensitivity_pushover_analysis(ctrlNode, baseNodes, dof, Dincr, max_disp,

˓→SensParam, IOflag=False):

85 """
86 Run pushover analysis with sensitivity
87 """
88 ops.wipeAnalysis()
89 start_time = time.time()
90 ops.loadConst("-time", 0.0)
91

92 title("Running Displacement-Control Pushover Sensitivity Analysis ...")

93

94 testType = "NormDispIncr" # EnergyIncr

95 tolInit = 1.0E-8
96 iterInit = 10 # Set the initial Max Number of Iterations
97 algorithmType = "Newton" # Set the algorithm type
98

99 ops.system("BandGeneral")
100 ops.constraints("Transformation")
101 ops.numberer("RCM")
102 ops.test(testType, tolInit, iterInit)
103 ops.algorithm(algorithmType)
104 # Change the integration scheme to be displacement control
105 # node dof init Jd min max
106 ops.integrator("DisplacementControl", ctrlNode, dof, Dincr)
107 ops.analysis("Static")
108 ops.sensitivityAlgorithm("-computeAtEachStep") # automatically compute
˓→sensitivity at the end of each step

109

110 if IOflag:
111 print(f"Single Pushover: Push node {ctrlNode} to {max_disp} {LunitTXT}.\n")
112

113 # Set some parameters

114 tCurrent = ops.getTime()
115 currentStep = 1
116

117 outputs = {"time": np.array([]),

118 "disp": np.array([]),
119 "force": np.array([]),
120 }
121

122 for sens in SensParam:

123 outputs[f"sensLambda_{sens}"] = np.array([]),
(continues on next page)

1.14. Examples 433

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

124

125 nodeList = []
126 for node in baseNodes:
127 nodeList.append(f"- ops.nodeReaction({node}, dof) ")
128

129 nodeList = "".join(nodeList)

130 currentDisp = ops.nodeDisp(ctrlNode, dof)
131 ok = 0
132

133 while ok == 0 and currentDisp < max_disp:

134 ops.reactions()
135 ok = ops.analyze(1)
136 tCurrent = ops.getTime()
137 currentDisp = ops.nodeDisp(ctrlNode, dof)
138

139 if IOflag:
140 print(f"Current displacement ==> {ops.nodeDisp(ctrlNode, dof):.3f}
˓→{LunitTXT}")

141

142 # if the analysis fails try initial tangent iteration

143 if ok != 0:
144 print("\n==> Trying relaxed convergence...")
145 ops.test(testType, tolInit/0.01, iterInit*50)
146 ok = ops.analyze(1)
147 if ok == 0:
148 print("==> that worked ... back to default analysis...\n")
149 ops.test(testType, tolInit, iterInit)
150

151 if ok != 0:
152 print("\n==> Trying Newton with initial then current...")
153 ops.test(testType, tolInit/0.01, iterInit*50)
154 ops.algorithm("Newton", "-initialThenCurrent")
155 ok = ops.analyze(1)
156 if ok == 0:
157 print("==> that worked ... back to default analysis...\n")
158 ops.algorithm(algorithmType)
159 ops.test(testType, tolInit, iterInit)
160

161 if ok != 0:
162 print("\n==> Trying ModifiedNewton with initial...")
163 ops.test(testType, tolInit/0.01, iterInit*50)
164 ops.algorithm("ModifiedNewton", "-initial")
165 ok = ops.analyze(1)
166 if ok == 0:
167 print("==> that worked ... back to default analysis...\n")
168 ops.algorithm(algorithmType)
169 ops.test(testType, tolInit, iterInit)
170

171 currentStep += 1
172 tCurrent = ops.getTime()
173

174 outputs["time"] = np.append(outputs["time"], tCurrent)

175 outputs["disp"] = np.append(outputs["disp"], ops.nodeDisp(ctrlNode, dof))
176 outputs["force"] = np.append(outputs["force"], eval(nodeList))
177

178 for sens in SensParam:

179 # sensLambda(patternTag, paramTag)
(continues on next page)

434 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

180 outputs[f"sensLambda_{sens}"] = np.append(outputs[f"sensLambda_{sens}"],
˓→ops.sensLambda(1, sens))

181

182 # Print a message to indicate if analysis completed succesfully or not

183 if ok == 0:
184 title("Pushover Analysis Completed Successfully!")
185 else:
186 title("Pushover Analysis Completed Failed!")
187

188 print(f"Analysis elapsed time is {(time.time() - start_time):.3f} seconds.\n")

189

190 return outputs

191

192

193 # +===============================================================================+
194 # | Define model |
195 # +===============================================================================+
196 # Create ModelBuilder
197 # -------------------
198 ops.wipe()
199 ops.model("basic", "-ndm", 2, "-ndf", 3)
200

201 # Create nodes

202 # ------------
203 ops.node(1, 0.0, 0.0) # Ground Level
204 ops.node(2, 30*ft, 0.0)
205 ops.node(3, 0.0, 15*ft) # 1st Floor Level
206 ops.node(4, 30*ft, 15*ft)
207 ops.node(5, 0.0, 2*15*ft) # 2nd Floor Level
208 ops.node(6, 30*ft, 2*15*ft)
209

210 # Fix supports at base of columns

211 # -------------------------------
212 ops.fix(1, 1, 1, 1)
213 ops.fix(2, 1, 1, 1)
214

215 # Define material

216 # ---------------
217 matTag = 1
218 Fy = 50.0*ksi # Yield stress
219 Es = 29000.0*ksi # Modulus of Elasticity of Steel
220 b = 1/100 # 1% Strain hardening ratio
221

222 # Sensitivity-ready steel materials: Hardening, Steel01, SteelMP, BoucWen, SteelBRB,

˓→StainlessECThermal, SteelECThermal, ...

223 ops.uniaxialMaterial("Steel01", matTag, Fy, Es, b)

224 # ops.uniaxialMaterial("SteelMP", matTag, Fy, Es, b)
225

226 # Define sections

227 # ---------------
228 # Sections defined with "canned" section ("WFSection2d"), otherwise use a
˓→FiberSection object (ops.section("Fiber",...))

229 colSecTag, beamSecTag = 1, 2

230 WSection = {"W18x76": [18.2*inch, 0.425*inch, 11.04*inch, 0.68*inch], # [d, tw, bf,
˓→tf]

231 "W14X90": [14.02*inch, 0.44*inch, 14.52*inch, 0.71*inch], # [d, tw, bf,

˓→tf]
(continues on next page)

1.14. Examples 435

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

232 }
233

234 # secTag, matTag, [d, tw, bf, tf], Nfw, Nff

235 ops.section("WFSection2d", colSecTag, matTag, *WSection["W14X90"], 20, 4) # Column
˓→section

236 ops.section("WFSection2d", beamSecTag, matTag, *WSection["W18x76"], 20, 4) # Beam

˓→section

237

238 # Define elements

239 # ---------------
240 colTransTag, beamTransTag = 1, 2
241 # Linear, PDelta, Corotational
242 ops.geomTransf("Corotational", colTransTag)
243 ops.geomTransf("Linear", beamTransTag)
244

245 colIntTag, beamIntTag = 1, 2

246 nip = 5
247 # Lobatto, Legendre, NewtonCotes, Radau, Trapezoidal, CompositeSimpson
248 ops.beamIntegration("Lobatto", colIntTag, colSecTag, nip),
249 ops.beamIntegration("Lobatto", beamIntTag, beamSecTag, nip)
250

251 # Column elements

252 ops.element("forceBeamColumn", 10, 1, 3, colTransTag, colIntTag, "-mass", 0.0)
253 ops.element("forceBeamColumn", 11, 3, 5, colTransTag, colIntTag, "-mass", 0.0)
254 ops.element("forceBeamColumn", 12, 2, 4, colTransTag, colIntTag, "-mass", 0.0)
255 ops.element("forceBeamColumn", 13, 4, 6, colTransTag, colIntTag, "-mass", 0.0)
256

257 # Beam elements

258 ops.element("forceBeamColumn", 14, 3, 4, beamTransTag, beamIntTag, "-mass", 0.0)
259 ops.element("forceBeamColumn", 15, 5, 6, beamTransTag, beamIntTag, "-mass", 0.0)
260

261 # Create a Plain load pattern with a Linear TimeSeries

262 # ----------------------------------------------------
263 ops.timeSeries("Linear", 1)
264 ops.pattern("Plain", 1, 1)
265

266 # +===============================================================================+
267 # | Define Sensitivity Parameters |
268 # +===============================================================================+
269 # /// Each parameter must be unique in the FE domain, and all parameter tags MUST be
˓→numbered sequentially starting from 1! ///

270 ops.parameter(1) # Blank parameters

271 ops.parameter(2)
272 ops.parameter(3)
273 for ele in [10, 11, 12, 13]: # Only column elements
274 ops.addToParameter(1, "element", ele, "E") # "E"
275 # Check the sensitivity parameter name in *.cpp files ("sigmaY" or "fy",
˓→somewhere also "Fy")

276 # https://github.com/OpenSees/OpenSees/blob/master/SRC/material/uniaxial/Steel02.
˓→cpp

277 ops.addToParameter(2, "element", ele, "fy") # "sigmaY" or "fy" or "Fy"

278 ops.addToParameter(3, "element", ele, "b") # "b"
279

280 title("Model Built")

281

282 # Run analysis with 10 steps

283 # -------------------------
(continues on next page)

436 Chapter 1. Developer

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

284 run_gravity_analysis(steps=10)
285

286 # +===============================================================================+
287 # | Define nodal loads & Run the analysis |
288 # +===============================================================================+
289 # Create nodal loads at nodes 3 & 5
290 # nd FX FY MZ
291 ops.load(3, 1/3, 0.0, 0.0)
292 ops.load(5, 2/3, 0.0, 0.0)
293

294 max_disp = 20*inch

295 pushover_output = run_sensitivity_pushover_analysis(
296 ctrlNode=5, baseNodes=[1, 2], dof=1, Dincr=(1/25)*inch, max_disp=max_disp,
˓→SensParam=ops.getParamTags(), IOflag=False)

297

298 # +===============================================================================+
299 # | Plot results |
300 # +===============================================================================+
301 rows, columns = len(ops.getParamTags())*2, 1
302 grid = plt.GridSpec(rows, columns, wspace=0.25, hspace=0.25)
303 fig = plt.figure(figsize=(10, 10))
304

305 ParamSym = ["E", "F_y", "b"]

306 ParamVars = [Es, Fy, b]
307

308

309 def plot_params():

310 plt.rcParams['axes.axisbelow'] = True
311 plt.xlim(xmin=0.0, xmax=max_disp)
312 plt.grid(True, color="silver", linestyle="solid", linewidth=0.75, alpha=0.75)
313 plt.tick_params(direction="in", length=5, colors="k", width=0.75)
314

315

316 for s in range(0, rows):

317 # Create subplots
318 # ---------------
319 if s < 3:
320 plt.subplot(grid[s]), plot_params()
321 plt.plot(pushover_output["disp"], pushover_output["force"], "-k", linewidth=2.
˓→0, label="$U$")

322 plt.plot(pushover_output["disp"], pushover_output["force"] +

323 pushover_output[f"sensLambda_{s+1}"] * ParamVars[s], "--k",
˓→linewidth=1.5, label=f"$U + (\partial V_b/\partial {ParamSym[s]}){ParamSym[s]}$")

324 plt.plot(pushover_output["disp"], pushover_output["force"] -

325 pushover_output[f"sensLambda_{s+1}"] * ParamVars[s], "-.k",
˓→linewidth=1.5, label=f"$U - (\partial V_b/\partial {ParamSym[s]}){ParamSym[s]}$")

326 plt.fill_between(pushover_output["disp"], pushover_output["force"] +

327 pushover_output[f"sensLambda_{s+1}"] * ParamVars[s],
˓→pushover_output["force"] - pushover_output[f"sensLambda_{s+1}"] * ParamVars[s],

˓→color='grey', alpha=0.15)

328 plt.ylabel(r"$V_b$ [kip]")

329 if s == 0:
330 plt.title(r"Displacement Control", fontsize=14)
331 if s == 1:
332 plt.ylabel(r"Base Shear, $V_b$ [kip]")
333 plt.legend(fontsize=9)
334
(continues on next page)

1.14. Examples 437

 OpenSeesPy Documentation, Release 3.5.1.3

(continued from previous page)

335 if s >= 3:
336 plt.subplot(grid[s]), plot_params()
337 plt.plot(pushover_output["disp"],
338 pushover_output[f"sensLambda_{s-2}"] * ParamVars[s-3], "-.k",
˓→linewidth=2.0, label="DDM")

339 plt.fill_between(pushover_output["disp"],
340 pushover_output[f"sensLambda_{s-2}"] * ParamVars[s-3], color=
˓→'grey', alpha=0.15)

341 plt.ylabel(f"$(\partial V_b/\partial {ParamSym[s-3]}){ParamSym[s-3]}$ [kip]")

342 if s == 5:
343 plt.xlabel(r"Roof Displacement, $U$ [in]")
344 plt.legend()
345

346 s += 1
347

348 # Save figure

349 # -----------
350 plt.savefig("SteelFrameSensitivity2D_v1.png", bbox_inches="tight", pad_inches=0.05,
˓→dpi=300, format="png")

351 plt.show()

438 Chapter 1. Developer

 Bibliography

[Boulanger-Ziotopoulou2018] R.W.Boulanger, K.Ziotopoulou. “PM4Silt(Version 1): A Silt Plasticity Model for

Earthquake Engineering Applications”. Report No. UCD/CGM-18/01 2018
[HintonEtAl1976] Hinton, E., Rock, T. & Zienkiewicz, O. (1976). “A note on mass lumping and related processes in
the Finite element method.” Earthquake Engineering and Structural Dynamics, 13, 9, p. A112.

439
OpenSeesPy Documentation, Release 3.5.1.3

440 Bibliography
 Index

A fix() (built-in function), 101

addToParameter() (built-in function), 288 fixX() (built-in function), 101
algorithm() (built-in function), 235 fixY() (built-in function), 101
analysis() (built-in function), 243 fixZ() (built-in function), 101
analyze() (built-in function), 244 frictionModel() (built-in function), 220

B G
barrier() (built-in function), 292 geomTransf() (built-in function), 222
basicDeformation() (built-in function), 257 getEleTags() (built-in function), 258
basicForce() (built-in function), 257 getLoadFactor() (built-in function), 259
basicStiffness() (built-in function), 257 getNodeTags() (built-in function), 259
Bcast() (built-in function), 293 getNP() (built-in function), 292
beamIntegration() (built-in function), 113 getNumThreads() (built-in function), 280
block2D() (built-in function), 112 getParamTags() (built-in function), 289
block3D() (built-in function), 112 getParamValue() (built-in function), 289
getPID() (built-in function), 292
C getTime() (built-in function), 259
computeGradients() (built-in function), 289 groundMotion() (built-in function), 110
constraints() (built-in function), 225
convertBinaryToText() (built-in function), 274
I
convertTextToBinary() (built-in function), 274 imposedMotion() (built-in function), 110
InitialStateAnalysis() (built-in function), 275
D integrator() (built-in function), 243
database() (built-in function), 275
domainChange() (built-in function), 294
L
layer() (built-in function), 213
E load() (built-in function), 107
eigen() (built-in function), 244 loadConst() (built-in function), 275
eleDynamicalForce() (built-in function), 258 logFile() (built-in function), 273
eleForce() (built-in function), 258
eleLoad() (built-in function), 108 M
element() (built-in function), 37 mass() (built-in function), 111
eleNodes() (built-in function), 258 mesh() (built-in function), 280
eleResponse() (built-in function), 258 modalDamping() (built-in function), 275
equalDOF() (built-in function), 102 modalProperties() (built-in function), 245
equalDOF_Mixed() (built-in function), 102 model() (built-in function), 37

F N
fiber() (built-in function), 212 nDMaterial() (built-in function), 190

441
OpenSeesPy Documentation, Release 3.5.1.3

node() (built-in function), 100 sectionStiffness() (built-in function), 272

nodeAccel() (built-in function), 259 sectionWeight() (built-in function), 272
nodeBounds() (built-in function), 259 send() (built-in function), 293
nodeCoord() (built-in function), 259 sensitivityAlgorithm() (built-in function), 289
nodeDisp() (built-in function), 260 sensLambda() (built-in function), 290
nodeDOFs() (built-in function), 260 sensNodeAccel() (built-in function), 290
nodeEigenvector() (built-in function), 260 sensNodeDisp() (built-in function), 289
nodeMass() (built-in function), 260 sensNodePressure() (built-in function), 290
nodePressure() (built-in function), 260 sensNodeVel() (built-in function), 290
nodeReaction() (built-in function), 261 sensSectionForce() (built-in function), 290
nodeResponse() (built-in function), 261 setElementRayleighDampingFactors()
nodeUnbalance() (built-in function), 261 (built-in function), 278
nodeVel() (built-in function), 261 setNodeAccel() (built-in function), 278
numberer() (built-in function), 226 setNodeCoord() (built-in function), 277
numFact() (built-in function), 262 setNodeDisp() (built-in function), 277
numIter() (built-in function), 262 setNodeVel() (built-in function), 278
setNumThreads() (built-in function), 279
P setParameter() (built-in function), 288
parameter() (built-in function), 287 setPrecision() (built-in function), 278
partition() (built-in function), 294 setStartNodeTag() (built-in function), 294
patch() (built-in function), 212 setTime() (built-in function), 277
pattern() (built-in function), 107 sp() (built-in function), 108
start() (built-in function), 278
preprocessing.DiscretizeMember.DiscretizeMember()
(built-in function), 295 stop() (built-in function), 278
pressureConstraint() (built-in function), 103 stripXML() (built-in function), 279
printA() (built-in function), 262 system() (built-in function), 228
printB() (built-in function), 262 systemSize() (built-in function), 273
printGID() (built-in function), 262
printModel() (built-in function), 262 T
test() (built-in function), 230
R testIter() (built-in function), 273
randomVariable() (built-in function), 291 testNorm() (built-in function), 273
rayleigh() (built-in function), 111 timeSeries() (built-in function), 103
reactions() (built-in function), 275
record() (built-in function), 263 U
recorder() (built-in function), 263 uniaxialMaterial() (built-in function), 119
recv() (built-in function), 293 updateElementDomain() (built-in function), 279
region() (built-in function), 111 updateParameter() (built-in function), 288
remesh() (built-in function), 285
remove() (built-in function), 276 V
reset() (built-in function), 276 version() (built-in function), 273
responseSpectrumAnalysis() (built-in func-
tion), 253 W
restore() (built-in function), 276
wipe() (built-in function), 279
rigidDiaphragm() (built-in function), 103
wipeAnalysis() (built-in function), 279
rigidLink() (built-in function), 103

S
save() (built-in function), 276
section() (built-in function), 210
sectionDeformation() (built-in function), 272
sectionFlexibility() (built-in function), 272
sectionForce() (built-in function), 271
sectionLocation() (built-in function), 272

442 Index