Enterprise Architect

User Guide Series

Parametric Simulation using

OpenModelica

Author: Sparx Systems

Date: 30/06/2017

Version: 1.0

CREATED WITH
Table of Contents

Parametric Simulation using OpenModelica 3

Install OpenModelica 4
Creating a Parametric Model 6
Configure SysML Simulation Window 17
Model Analysis using Datasets 21
SysML Simulation Examples 23
Electrical Circuit Simulation Example 24
Mass-Spring-Damper Oscillator Simulation Example 31
Water Tank Pressure Regulator 38
Troubleshooting OpenModelica Simulation 47
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Parametric Simulation using OpenModelica

Enterprise Architect provides integration with OpenModelica to support rapid and robust evaluation of how a SysML
model will behave in different circumstances.
This section describes the process of defining a Parametric model, annotating the model with additional information to
drive a simulation, and running a simulation to generate a graph.

Introduction to SysML Parametric Models

SysML Parametric models support the engineering analysis of critical system parameters, including the evaluation of key
metrics such as performance, reliability and other physical characteristics. These models combine requirements models
with system design models, by capturing executable constraints based on complex mathematical relationships.
Parametric diagrams are specialized Internal Block diagrams that help you, the modeler, to combine behavior and
structure models with engineering analysis models such as performance, reliability, and mass property models.
For further information on the concepts of SysML Parametric models, refer to the official OMG SysML website and its
linked sources.

SysMLSimConfiguration Artifact

Enterprise Architect helps you to extend the usefulness of your SysML parametric models by annotating them with extra
information that allows the model to be simulated. The resulting model is then generated as a Modelica model that can be
solved (simulated) using OpenModelica.
The simulation properties for your model are stored against a Simulation Artifact. This preserves your original model and
supports multiple simulations being configured against a single SysML model. The Simulation Artifact can be found on
the 'Artifacts' Toolbox page.

User Interface

The user interface for the SysML simulation is described in the Configure SysML Simulation Window topic.

OpenModelica Examples

To aid your understanding of how to create and simulate a SysML parametric model, three examples have been provided
to illustrate three different domains. These examples and what you are able to learn from them are described in the
SysML Simulation Examples topic.

(c) Sparx Systems 2015 - 2017 Page 3 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Install OpenModelica
On Windows

Download the OpenModelica Installer from https://openmodelica.org/download/download-windows

Double-click on the OpenModelica installer and follow the wizards
Make sure you can locate omc.exe (for example, C:\OpenModelica1.9.2\bin\omc.exe)

On Linux

Go to the URL (https://openmodelica.org/download/download-linux) and follow the instructions:

1. Run this script to add OpenModelica to your additional repository list

for deb in deb deb-src; do echo "$deb http://build.openmodelica.org/apt `lsb_release -cs` release"; done | sudo tee
/etc/apt/sources.list.d/openmodelica.list

Note: If you are installing on Linux Mint Rosa, this Repository can be created:
deb http://build.openmodelica.org/apt rosa nightly
deb-src http://build.openmodelica.org/apt rosa nightly

You must change 'rosa' to 'trusty' in order to make it work.

Menu | Search Bar | Software Sources (type in password) | Additional repositories | Select 'Openmodelica' | Edit URL |
change rosa to trusty | OK | Do the same for 'Openmodelica(Sources)'

2. Import the GPG key used to sign the releases:

wget -q http://build.openmodelica.org/apt/openmodelica.asc -O- | sudo apt-key add -

3. Update and install OpenModelica

sudo apt-get update
sudo apt-get install openmodelica
sudo apt-get install omlib-.* # Installs optional Modelica libraries (most have not been tested with OpenModelica)

Check that you can find the file under /usr/bin/omc.

Configure OpenModelica in Enterprise Architect

Method Select

Ribbon Simulate > SysMLSim > Manage > SysMLSim Configuration Manager > Menu >
Configure Modelica Solver

Double-click on an Artifact with the SysMLSimConfiguration stereotype > Menu >

(c) Sparx Systems 2015 - 2017 Page 4 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Other Configure Modelica Solver

Configure Solver Display the 'Modelica Solver Path' dialog, in which you type or browse for the path
to the Modelica solver to use.
For windows, it looks like this:

For Linux, it looks like this:

(c) Sparx Systems 2015 - 2017 Page 5 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Creating a Parametric Model

In this topic we discuss how you might develop SysML model elements for simulation (assuming existing knowledge of
SysML modeling), configure these elements in the Configure SysML Simulation window, and observe the results of a
simulation under some of the different definitions and modeling approaches. The points are illustrated by snapshots of
diagrams and screens from the SysML Simulation examples provided in this chapter.
When creating a Parametric Model, you can apply one of three approaches to defining Constraint Equations:
Defining inline Constraint Equations on a Block element
Creating re-usable Constraint Blocks, and
Using connected constraint properties
You would also take into consideration:
Flows in physical interactions
Default Values and Initial Values
Simulation Functions
Value Allocation, and
Packages and Imports

Access

Ribbon Simulate > SysMLSim > Manage > SysMLSim Configuration Manager

Defining inline Constraint Equations on a Block

Defining constraints directly in a Block is straightforward and is the easiest way to define constraint equations.
In this figure, constraint 'f = m * a' is defined in a Block element.

block
FMA_Test
constraints
{f=m*a}
properties
a = 9.81
f
m = 10

Tip: You can define multiple constraints in one Block.

1. Create a SysMLSim Configuration Artifact 'Force=Mass*Acceleration(1)' and point it to the Package 'FMA_Test'.
2. For 'FMA_Test', in the 'Value' column set 'SysMLSimModel'.
3. For Parts 'a', 'm' and 'f', in the 'Value' column set 'a' and 'm' to 'SimConstant' and (optionally) set 'f' to 'SimVariable'.
4. On the 'Simulation' tab, in the 'Properties to Plot' panel, select the checkbox against 'f'.
5. Click on the Solve button to run the simulation.

(c) Sparx Systems 2015 - 2017 Page 6 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

A chart should be plotted with f = 98.1 (which comes from 10 * 9.81).

Connected Constraint Properties

In SysML, constraint properties existing in Constraint Blocks can be used to provide greater flexibility in defining
constraints.
In this figure, Constraint Block 'K' defines parameters 'a', 'b', 'c', 'd' and 'KVal', and three constraint properties 'eq1', 'eq2'
and 'eq3', typed to 'K1', 'K2' and 'K1MultiplyK2' respectively.

constraint
K
constraint properties
eq1 : K1
eq2 : K2
eq3 : K1MultiplyK2
parameters
a
b
c
d
KVal

+eq1 +eq3 +eq2

constraint constraint constraint

K1 K1MultiplyK2 K2

constraints constraints constraints

{K1 = x * y} {K=K1*K2} {p = K2 / q}

parameters parameters parameters

x K K2
K1 K1 p
y K2 q

Create a Parametric diagram in Constraint Block 'K' and connect the parameters to the constraint properties with Binding

(c) Sparx Systems 2015 - 2017 Page 7 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

connectors, as shown:
par [constraint block] K [K]

equal
p c
a x eq1 : K1 eq2 : K2 equal
K1 K2 {p = K2 / q}
{K1 = x * y}
b y q d
equal equal

equal equal

K1 K2
eq3 : K1MultiplyK2
{K=K1*K2}

equal

KVal

Create a model MyBlock with five Properties (Parts)

Create a constraint property 'eq' for MyBlock and show the parameters
Bind the properties to the parameters

par [block] MyBlock [MyBlockPar]

arg_a equal a eq : K

arg_b equal b
KVal arg_K
equal
arg_c equal c

arg_d d
equal

Provide values (arg_a = 2, arg_b = 3, arg_c = 4, arg_d = 5) in a data set

In the 'Configure SysML Simulation' dialog, set 'Model' to 'MyBlock' and 'Data Set' to 'DataSet_1'
In the 'Properties to Plot' panel, select the checkbox against 'arg_K'
Click on the Solve button to run the simulation

(c) Sparx Systems 2015 - 2017 Page 8 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

The result 120 (calculated as 2 * 3 * 4 * 5) will be computed and plotted. This is the same as when we do an expansion
with pen and paper: K = K1 * K2 = (x*y) * (p*q), then bind with the values (2 * 3) * (4 * 5); we get 120.
What is interesting here is that we intentionally define K2's equation to be 'p = K2 / q' and this example still works.
We can easily solve K2 to be p * q in this example, but in some complex examples it is extremely hard to solve a
variable from an equation; however, the Enterprise Architect SysMLSim can still get it right.
In summary, the example shows you how to define a Constraint Block with greater flexibility by constructing the
constraint properties. Although we demonstrated only one layer down into the Constraint Block, this mechanism could
work on complex models for an arbitrary level of use.

Creating Reuseable Constraint Blocks

If one equation is commonly used in many Blocks, a Constraint Block can be created for use as a constraint property in
each Block. These are the changes we make, based on the previous example:
Create a Constraint Block element 'F_Formula' with three parameters 'a', 'm' and 'f', and a constraint 'f = m * a'

Tip: Primitive type 'Real' will be applied if property types are empty

(c) Sparx Systems 2015 - 2017 Page 9 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Create a Block 'FMA_Test' with three properties 'x', 'y' and 'z', and give 'x' and 'y' the default values '10' and '9.81'
respectively
Create a Parametric diagram in 'FMA_Test', showing the properties 'x', 'y' and 'z'
Create a Constraint Property 'e1' typed to 'F_Formula' and show the parameters
Draw Binding connectors between 'xm', 'ya', and 'fz' as shown:

par [block] FMA_Test [testingFormulaF]

e1 : F_Formula
{f=m*a}
x m
equal f z
equal
y a
equal

Create a SysMLSimConfiguration Artifact

element and configure it as shown in the dialog illustration:
- In the 'Value' column, set 'FMA_Test' to 'SysMLSimModel'
- In the 'Value' column, set 'x' and 'y' to 'SimConstant'
- In the 'Properties to Plot' panel select the checkbox against 'Z'
- Click on the Solve button to run the simulation

A chart should be plotted with f = 98.1 (which comes from 10 * 9.81).

Flows in Physical Interactions

When modeling for physical interaction, exchanges of conserved physical substances such as electrical current, force,
torque and flow rate should be modeled as flows, and the flow variables should be set to the attribute 'isConserved'.
Two different types of coupling are established by connections, depending on whether the flow properties are potential

(c) Sparx Systems 2015 - 2017 Page 10 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

(default) or flow (conserved):

Equality coupling, for potential (also called effort) properties
Sum-to-zero coupling, for flow (conserved) properties; for example, according to Kirchoff's Current Law in the
electrical domain, conservation of charge makes all charge flows into a point sum to zero

In the generated Modelica code of the 'ElectricalCircuit' example:

connector ChargePort
flow Current i; //flow keyword will be generated if 'isConserved' = true
Voltage v;
end ChargePort;

model Circuit
Source source;
Resistor resistor;
Ground ground;
equation
connect(source.p, resistor.n);
connect(ground.p, source.n);
connect(resistor.p, source.n);
end Circuit;

Each connect equation is actually expanded to two equations (there are two properties defined in ChargePort), one for
equality coupling, the other for sum-to-zero coupling:
source.p.v = resistor.n.v;
source.p.i + resistor.n.i = 0;

Default Value and Initial Values

If initial values are defined in SysML property elements ('Properties' dialog > 'Property' page > 'Initial' field), they can be
loaded as the default value for a SimConstant or the initial value for a SimVariable.
In this Pendulum example, we have provided initial values for properties 'g', 'L', 'm', 'PI', 'x' and 'y', as seen on the left
hand side of the figure. Since 'PI' (the mathematical constant), 'm' (mass of the Pendulum), 'g' (Gravity factor) and 'L'
(Length of Pendulum) do not change during simulation, set them as 'SimConstant'.

(c) Sparx Systems 2015 - 2017 Page 11 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

block
Pendulum This example is a mathematical
properties model of a physical system.
F
g : External Reference = 9.81
L = 0.5 The equations are Newton's
m : External Reference = 1
PI : External Reference = 3.141 equations of motion for the
vx : External Reference
vy : External Reference
pendulum mass under the
x : External Reference = 0.5 influence of gravity.
y : External Reference = 0

constraints
e_newton_x : Newton_pendulum_balance_x
e_newton_y : Newton_pendulum_balance_y
eRightTrangle : RightTriangle
ex : SimpleDer
ey : SimpleDer

The generated modelica code resembles this:

class Pendulum
parameter Real PI = 3.141;
parameter Real m = 1;
parameter Real g = 9.81;
parameter Real L = 0.5;
Real F;
Real x (start=0.5);
Real y (start=0);
Real vx;

(c) Sparx Systems 2015 - 2017 Page 12 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Real vy;
......
equation
......
end Pendulum;

Properties 'PI', 'm', 'g' and 'L' are constant, and are generated as a declaration equation
Properties 'x' and 'y' are variable; their starting values are 0.5 and 0 respectively, and the initial values are generated
as modifications

Simulation Functions

A Simulation function is a powerful tool for writing complex logic, and is easy to use for constraints. This section
describes a function from the TankPI example.
In the Constraint Block 'Q_OutFlow', a function 'LimitValue' is defined and used in the constraint.

constraint
Q_OutFlow

SimFunction
+ LimitValue(double, double, double, *double): int
constraints
{a=LimitValue(min, max, -b*c)}
parameters
a
b
c
max
min

On a Block or Constraint Block, create an operation ('LimitValue' in this example) and open the 'Operations' tab of
the 'Features' dialog
Give the operation the stereotype 'SimFunction'
Define the parameters and set the direction to 'in/out'

Tips: Multiple parameters could be defined as 'out', and the caller retrieves the value in format of:
(out1, out2, out3) = function_name(in1, in2, in3, in4, ...); //Equation form
(out1, out2, out3) := function_name(in1, in2, in3, in4, ...); //Statement form

Define the function body in the 'Initial Code' field of the 'Behavior' tab, as shown:

pLim :=
if p > pMax then
pMax
else if p < pMin then
pMin
else
p;

(c) Sparx Systems 2015 - 2017 Page 13 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

When generating code, Enterprise Architect will collect all the operations stereotyped as 'SimFunction' defined in
Constraint Blocks and Blocks, then generate code resembling this:
function LimitValue
input Real pMin;
input Real pMax;
input Real p;
output Real pLim;
algorithm
pLim :=
if p > pMax then
pMax
else if p < pMin then
pMin
else
p;
end LimitValue;

Value Allocation

This figure shows a simple model called 'Force=Mass*Acceleration'.

constraint block block

F_Formula FMA FMA_Test
constraints properties constraints
{f=m*a} a {a_value=sin(time)}
f {m_value=cos(time)}
parameters m
a properties
f constraints a_value
m e1 : F_Formula fma1 : FMA
m_value

A block 'FMA' is modeled with properties 'a', 'f', and 'm' and a constraintProperty 'e1', typed to Constraint Block
'F_Formula'
The block 'FMA' does not have any initial value set on its properties, and the properties 'a', 'f' and 'm' are all variable,
so their value change depends on the environment in which they are simulated
Create a block 'FMA_Test' as a SysMLSimModel and add the property 'fma1' to test the behavior of block 'FMA'
Constraint 'a_value' to be 'sin(time)'
Constraint 'm_value' to be 'cos(time)'
Draw Allocation connectors to allocate values from environment to the model 'FMA'

(c) Sparx Systems 2015 - 2017 Page 14 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

ibd [block] FMA_Test [FMA_Test]

fma1: FMA

value constraint as a
"sin(time)" a_value
a llocate

value constraint as
m m_value
a llocate "cos(time)"

Select the 'Properties to Plot' checkboxes against 'fma1.a', 'fma1.m' and 'fma1.f'
Click on the Solve button to simulate the model

Packages and Imports

The SysMLSimConfiguration Artifact collects the elements (such as Blocks, Constraint Blocks and Value Types) of a
Package. If the simulation depends on elements not owned by this Package, such as Reusable libraries, Enterprise
Architect provides an Import connector between Package elements to meet this requirement.
In the Electrical Circuit example, the Artifact is configured to the Package 'ElectricalCircuit', which contains almost all
of the elements needed for simulation. However, some properties are typed to value types such as 'Voltage', 'Current' and
'Resistance', which are commonly used in multiple SysML models and are therefore placed in a Package called
'CommonlyUsedTypes' outside the individual SysML models. If you import this Package using an Import connector, all
the elements in the imported Package will appear in the SysMLSim Configuration Manager.

ElectricalCircuit
+ ElectricalCircuit
+ ChargePort
+ Circuit
CommonlyUsedTypes
+ Ground
+ GroundConstraint + Current
+ Resistor import + Resistance
+ ResistorConstraint + Voltage
+ Source
+ SourceConstraint (from Modelica Examples)
+ TwoPinComponent
+ TwoPinComponentConstraint

(from Modelica Examples)

(c) Sparx Systems 2015 - 2017 Page 15 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

(c) Sparx Systems 2015 - 2017 Page 16 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Configure SysML Simulation Window

The Configure SysML Simulation window is the interface through which you can provide run-time parameters for
executing the simulation of a SysML model. The simulation is based on a simulation configuration defined in a
SysMLSimConfiguration Artifact element.

Access

Ribbon Simulate > SysMLSim > Manage > SysMLSim Configuration Manager

Other Double-click on an Artifact with the SysMLSimConfiguration stereotype.

Toolbar Options

Option Description

Click on the drop-down arrow and select from these options:

Select Artifact - Select and load an existing configuration from an Artifact with
the SysMLSimConfiguration stereotype (if one has not already been selected)
Create Artifact - Create a new SysMLSimConfiguration or select and load an
existing configuration artifact
Select Package - Select a Package to scan for SysML elements to configure for
simulation
Reload - Reload the Configuration Manager with changes to the current
Package
Configure Modelica Solver - Display the 'Modelica Solver Path' dialog, in
which you type or browse for the path to the Modelica solver to use

(c) Sparx Systems 2015 - 2017 Page 17 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Click on this button to save the configuration to the current Artifact.

Click on this button to generate, compile and run the current configuration, and
display the results.

After simulation, the result file is generated in either plt, mat or csv format. That is,
with the filename:
ModelName_res.plt
ModelName_res.mat or
ModelName_res.csv
Click on this button to specify a directory into which Enterprise Architect will copy
the result file.

Click on this button to select from these options:

Generate Modelica Code - Generate the code without compiling or running it
Open Modelica Simulation Directory - Open the directory into which Modelica
code will be generated
Edit Modelica Templates - Customize the code generated for Modelica, using
the Code Template Editor

Simulation Artifact and Model Selection

Field Action

Artifact
Click on the icon and either browse for and select an existing
SysMLSimConfiguration Artifact, or create a new Artifact.

Package If you have specified an existing SysMLSimConfiguration Artifact, this field

defaults to the Package containing the SysML model associated with that Artifact.

Otherwise, click on the icon and browse for and select the Package containing
the SysML model to configure for simulation. You must specify (or create) the
Artifact before selecting the Package.

Package Objects

(c) Sparx Systems 2015 - 2017 Page 18 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

This table discusses the types of object from the SysML model that will be listed under the 'Name' column in the
Configure SysML Simulation window, to be processed in the simulation. Each object type expands to list the named
objects of that type, and the properties of each object that require configuration in the 'Value' column.
Many levels of the object types, names and properties do not require configuration, so the corresponding 'Value' field
does not accept input. Where input is appropriate and accepted, a drop-down arrow displays at the right end of the field;
when you click on this a short list of possible values displays for selection. Certain values (such as 'SimVariable' for a
Part) add further layers of parameters and properties, where you click on the button to, again, select and set values
for the parameters. For datasets, the input dialog allows you to type in or import values, such as initial or default values;
see the Model Analysis using Datasets topic.

Element Type Behavior

ValueType ValueType elements either generalize from a primitive type or are substituted by
SysMLSimReal for simulation.

Block Block elements mapped to SysMLSimClass or SysMLSimModel elements support

the creation of data sets. If you have defined multiple data sets in a SysMLSimClass
(which can be generalized), you must identify one of them as the default (using the
context menu option 'Set as Default Dataset').
As a SysMLSimModel is a possible top level element for a simulation, and will not
be generalized, if you have defined multiple datasets the dataset to use is chosen
during the simulation.

Properties Properties within a Block can be configured to be either SimConstants or

SimVariables. For a SimVariable, you configure these attributes:
isContinuous - determines whether the property value varies continuously
('true', the default) or discretely ('false')
isConserved - determines whether values of the property are conserved ('true')
or not ('false', the default); when modeling for physical interaction, the
interactions include exchanges of conserved physical substances such as
electrical current, force or fluid flow
changeCycle - specifies the time interval at which a discrete property value
changes; the default value is '0'
- changeCycle can be set to a value other than 0 only when isContinuous =
'false'
- The value of changeCycle must be positive or equal to 0

Port No configuration required.

SimFunction Functions are created as operations in Blocks or Constraint Blocks, stereotyped as

'SimFunction'.
No configuration is required in the Configure SysML Simulation window.

Generalization No configuration required.

Binding Connector Binds a property to a parameter of a constraint property.

No configuration required.

Connector Connects two Ports.

No configuration required in the Configure SysML Simulation window. However,
you might have to configure the properties of the Port's type by determining
whether the attribute isConserved should be set as 'False' (for potential properties,
so that equality coupling is established) or 'True' (for flow/conserved properties, so
that sum-to-zero coupling is established).

(c) Sparx Systems 2015 - 2017 Page 19 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Constraint Block No configuration required.

Simulation Tab

This table describes the fields of the 'Simulation' tab on the Configure SysML Simulation window.

Field Action

Model Click on the drop-down arrow and select the top level node (a SysMLSimModel
element) for the simulation. The list is populated with the names of the Blocks
defined as top-level, model nodes.

Data Set Click on the drop-down arrow and select the dataset for the selected model.

Start Type in the initial wait time before which the simulation is started, in seconds
(default value is 0).

Stop Type in the number of seconds for which the simulation will execute.

Format Click on the drop-down arrow and select either 'plt', 'csv' or 'mat' as the format of
the result file, which could potentially be used by other tools.

Parametric Plot Select this checkbox to plot Legend A on the y-axis against Legend B on the
x-axis.
Deselect the checkbox to plot Legend(s) on the y-axis against time on the
x-axis

Note: With the checkbox selected, you must select two properties to plot.

Dependencies Lists the types that must be generated to simulate this model.

Properties to Plot Provides a list of variable properties that are involved with the simulation. Select
the checkbox against each property to plot.

(c) Sparx Systems 2015 - 2017 Page 20 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Model Analysis using Datasets

Every SysML Block used in a Parametric model can, within the Simulation configuration, have multiple datasets defined
against it. This allows for repeatable simulation variations using the same SysML model.
A Block can be typed as a SysMLSimModel (a top-level node that cannot be generalized or form part of a composition)
or as a SysMLSimClass (a lower-level element that can be generalized or form part of a composition). When running a
simulation on a SysMLSimModel element, if you have defined multiple datasets, you can specify which dataset to use.
However, if a SysMLSimClass within the simulation has multiple datasets, you cannot select which one to use during the
simulation and must therefore identify one dataset as the default for that Class.

Access

Ribbon Simulate > SysMLSim > Manage > SysMLSim Configuration Manager > in
"block" group > Name column > Context menu on block element > Create
Simulation DataSet

Dataset Management

Task Action

Create To create a new dataset, right-click on a Block name and select the 'Create
Simulation Dataset' option. The dataset is added to the end of the list of components
underneath the Block name. Click on the button to set up the dataset on the
'Configure Simulation Data' dialog (see the Configure Simulation Data table).

Duplicate To duplicate an existing dataset as a base for creating a new dataset, right-click on
the dataset name and select the 'Duplicate' option. The duplicate dataset is added to
the end of the list of components underneath the Block name. Click on the
button to edit the dataset on the 'Configure Simulation Data' dialog (see the
Configure Simulation Data table).

Delete To remove a dataset that is no longer required, right-click on the dataset and select
the 'Delete Dataset' option.

Set Default To set the default dataset used by a SysMLSimClass when used as a property type
or inherited (and when there is more than one dataset), right-click on the dataset
and select the 'Set as Default' option. The name of the default dataset is highlighted
in bold. The properties used by a model will use this default configuration unless
the model overrides them explicitly.

Configure Simulation Data

This dialog is principally for information. The only column in which you can directly add or change data is the 'Value'
column.

(c) Sparx Systems 2015 - 2017 Page 21 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Column Description

Attribute The 'Attribute' column provides a tree view of all the properties in the Block being
edited.

Stereotype The 'Stereotype' column identifies, for each property, if it has been configured to be
a constant for the duration of the simulation or variable, so that the value is
expected to change over time.

Type The 'Type' column describes the type used for simulation of this property. It can be
either a primitive type (such as 'Real') or a reference to a Block contained in the
model. Properties referencing Blocks will show the child properties specified by the
referenced Block below them.

Default Value The 'Default Value' column shows the value that will be used in the simulation if no
override is provided. This can come from the 'Initial Value' field in the SysML
model or from the default dataset of the parent type.

Value The 'Value' column allows you to override the default value for each primitive
value.

Export / Import Click on these buttons to modify the values in the current dataset using an external
application - such as a spreadsheet - before re-importing them.

(c) Sparx Systems 2015 - 2017 Page 22 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

SysML Simulation Examples

This section provides a worked example for each of: creating a SysML model for a domain, simulating it, and evaluating
the results of the simulation. These examples apply the information discussed in the earlier topics.

Electrical Circuit Simulation Example

The first example is of the simulation of an electrical circuit. The example starts with an electrical circuit diagram and
converts it to a parametric model. The model is then simulated and the voltage at the source and target wires of a resistor
are evaluated and compared to the expected values.
Electrical Circuit Simulation Example

Mass-Spring-Damper Oscillator Simulation Example

The second example uses a simple physical model to demonstrate the oscillation behavior of a string under tension.
Mass-Spring-Damper Oscillator Simulation Example

Water Tank Pressure Regulator

The final example shows the water levels of two water tanks where the water is being distributed between them. We first
simulate a well balanced system, then we simulate a system where the water will overflow from the second tank.
Water Tank Pressure Regulator

(c) Sparx Systems 2015 - 2017 Page 23 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Electrical Circuit Simulation Example

In this section, we will walk through the creation of a SysML parametric model for a simple electrical circuit, and then
use a parametric simulation to predict and chart the behavior of that circuit.

Circuit Diagram

The electrical circuit we are going to model, shown here, uses a standard electrical circuit notation.

The circuit includes an AC power source, a ground and a resistor, connected to each other by wires.

Create SysML Model

This table shows how we can build up a complete SysML model to represent the circuit, starting at the lowest level types
and building up the model one step at a time.

Component Action

Types Define Value Types for the Voltage, Current and Resistance. Unit and quantity kind
are not important for the purposes of simulation, but would be set if defining a
complete SysML model. These types will be generalized from the primitive type
'Real'. In other models, you can choose to map a Value Type to a corresponding
simulation type separate from the model.

valueType valueType valueType

Voltage Current Resistance

quantityKind = quantityKind = quantityKind =

unit = unit = unit =

Additionally, define a composite type called ChargePort, which includes properties

for both Current and Voltage. This type allows us to represent the electrical energy
at the connectors between components.

(c) Sparx Systems 2015 - 2017 Page 24 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

block
ChargePort
flow properties
none i : Current
none v : Voltage

Blocks In SysML, the circuit and each of the components will be represented as Blocks.
Create a Circuit Block in a Block Definition Diagram (BDD). The circuit has three
parts: a source, a ground, and a resistor. These parts are of different types, with
different behaviors. Create a Block for each of these part types. The three parts of
the Circuit Block are connected through Ports, which represent electrical pins. The
source and resistor have a positive and a negative pin. The ground has only one pin,
which is positive. Electricity (electric charge) is transmitted through the pins.
Create an abstract block 'TwoPinComponent' with two Ports (pins). The two Ports
are named 'p' (positive) and 'n' (negative), and they are of type ChargePort.
This figure shows what the BDD should look like, with the blocks Circuit, Ground,
TwoPinComponent, Source and Resistor.

block block block

Ground p: ChargePort TwoPinComponent n: ChargePort
Circuit

properties ports ports

resistor : Resistor p : ChargePort p : ChargePort
ground : Ground n : ChargePort
constraints
source : Source
gc : GroundConstraint values
v : Voltage
p: ChargePort i : Current

block block
Source Resistor

ports ports
p : ChargePort n : ChargePort
n : ChargePort p : ChargePort

values values
i : Current i : Current
v : Voltage r : Resistance = 10
v : Voltage
constraints
sc : SourceConstraint constraints
rc : ResistorConstraint

Internal Structure Create an Internal Block Diagram (IBD) for Circuit. Add properties for the Source,
Resistor and Ground, typed by the corresponding Blocks. Connect the Ports with
connectors. The positive pin of the Source is connected to the negative pin of the
Resistor. The positive pin of the Resistor is connected to the negative pin of the
Source. The Ground is also connected to the negative pin of the Source.

(c) Sparx Systems 2015 - 2017 Page 25 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

ibd [block] Circuit [Circuit]

p: ChargePort n: ChargePort

source: Source resistor: Resistor

n: ChargePort
p: ChargePort

p: ChargePort

ground: Ground

Notice that this follows the same structure as the original circuit diagram, but the
symbols for each component have been replaced with properties typed by the
Blocks we have defined.

Constraints Equations define mathematical relationships between numeric properties. In

SysML, equations are represented as constraints in Constraint Blocks. Parameters
of Constraint Blocks correspond to SimVariables and SimConstants of Blocks ('i',
'v', 'r' in this example), as well as to SimVariables present in the type of the Ports
('pv', 'pi', 'nv', 'ni' in this example).
Create a Constraint Block 'TwoPinComponentConstraint' to define parameters and
equations common to sources and resistors. The equations should state that the
voltage of the component is equal to the difference between the voltages at the
positive and negative pins. The current of the component is equal to the current
going through the positive pin. The sum of the currents going through the two pins
must add up to zero (one is the negative of the other). The Ground constraint states
that the voltage at the Ground pin is zero. The Source constraint defines the voltage
as a sine wave with the current simulation time as a parameter. This figure shows
what these constraints should look like in a BDD.

(c) Sparx Systems 2015 - 2017 Page 26 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

constraint constraint
GroundConstraint TwoPinComponentConstraint

constraints constraints
{pv=0} {pi+ni=0}
{i=pi}
parameters {v=pv-nv}
pv : Real
parameters
i : Real
ni : Real
nv : Real
pi : Real
pv : Real
v : Real

constraint
constraint
ResistorConstraint
SourceConstraint
constraints
constraints
{v=r*i}
{v=sin(time)}
parameters
parameters
i : Real
i : Real
ni : Real
ni : Real
nv : Real
nv : Real
pi : Real
pi : Real
pv : Real
pv : Real
r : Real
v : Real
v : Real

Bindings The values of Constraint parameters are equated to variable and constant values
with binding connectors. Create Constraint properties on each Block (properties
typed by Constraint Blocks) and bind the Block variables and constants to the
Constraint parameters to apply the Constraint to the Block. These figures show the
bindings for the Ground, the Source and the Resistor respectively.
For the Ground constraint, bind gc.pv to p.v.
par [block] Ground [Ground]

gc : GroundConstraint
{pv=0}
v : Voltage
equal pv : Real

p: ChargePort

For the Source constraint, bind:

sc.pi to p.i
sc.pv to p.v
sc.v to v
sc.i to i
sc.ni to n.i and
sc.nv to n.v

(c) Sparx Systems 2015 - 2017 Page 27 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

par [block] Source [Source]

sc : SourceConstraint
{v=sin(time)}
i : Current pi : Real ni : Real i : Current
equal equal

nv : Real
v : Voltage pv : Real equal v : Voltage
equal

p: ChargePort i : Real v : Real n: ChargePort

equal
equal

i: Current v: Voltage

For the Resistor constraint, bind:

rc.pi to p.i
rc.pv to p.v
rc.v to v
rc.i to i
rc.ni to n.i
rc.nv to n.v and
rc.r to r
par [block] Resistor [Resistor]

rc : ResistorConstraint ni : Real
i : Current pi : Real {v=r*i} equal i : Current
equal

v : Voltage pv : Real nv : Real v : Voltage

equal equal

p:
p: ChargePort
ChargePort n:
n: ChargePort
ChargePort
v : Real i : Real r : Real
equal equal equal

v: Voltage i: Current r: Resistance

Configure Simulation Behavior

This table shows the detailed steps of the configuration of SysMLSim.

Step Action

SysMLSimConfiguration Select 'Simulate > SysMLSim > Manage > SysMLSim Configuration Manager'
Artifact From the first toolbar icon drop-down, select 'Create Artifact' and create the
Artifact element
Select the Package that owns this SysML Model

Create Root elements in

(c) Sparx Systems 2015 - 2017 Page 28 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Configuration Manager ValueType

block
constraintBlock

ValueType Substitution Expand ValueType and for each of Current, Resistance and Voltage select
'SysMLSimReal' from the 'Value' combo box.

Set property as flow Expand 'block' to ChargePort | FlowProperty | i : Current and select
'SimVariable' from the 'Value' combo box

For 'SysMLSimConfiguration' click on the button to open the 'Element

Configurations' dialog
Set 'isConserved' to 'True'

SysMLSimModel This is the model we want to simulate: set the block 'Circuit' to be
'SysMLSimModel'.

Run Simulation

In the 'Simulation' page, select the checkboxes against 'resistor.n.v' and 'resistor.p.v' for plotting and click on the Solve
button.

The two legends 'resistor.n.v' and 'resistor.p.v' are plotted, as shown.

(c) Sparx Systems 2015 - 2017 Page 29 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

(c) Sparx Systems 2015 - 2017 Page 30 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Mass-Spring-Damper Oscillator Simulation Example

In this section, we will walk through the creation of a SysML parametric model for a simple Oscillator composed of a
mass, a spring and a damper, and then use a parametric simulation to predict and chart the behavior of this mechanical
system. Finally, we perform what-if analysis by comparing two oscillators provided with different parameter values
through data sets.

System being modeled

A mass is hanging on a spring and damper. The first state shown here represents the initial point at time=0, just when the
mass is released. The second state represents the final point when the body is at rest and the spring forces are in
equilibrium with gravity.

Create SysML Model

The MassSpringDamperOscillator model in SysML has a main Block, the Oscillator. The Oscillator has four parts: a
fixed ceiling, a spring, a damper and a mass body. Create a Block for each of these parts. The four parts of the Oscillator
Block are connected through Ports, which represent mechanical flanges.

Components Description

Port Types The Blocks 'Flange_a' and 'Flange_b' used for flanges in the 1D translational
mechanical domain are identical but have slightly different roles, somewhat
analogous to the roles of PositivePin and NegativePin in the electrical domain.
Momentum is transmitted through the flanges. So the attribute isConserved of flow
property Flange.f should be set to True.

(c) Sparx Systems 2015 - 2017 Page 31 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

block
Flange

flow properties
inout f
inout s

block block
Flange_a Flange_b

Blocks and Ports Create Blocks 'Spring', 'Damper', 'Mass' and 'Fixed' to represent the spring,
damper, mass body and ceiling respectively
Create a Block 'PartialCompliant' with two Ports (flanges), named 'flange_a'
and 'flange_b' - these are of type Flange_a and Flange_b respectively; the
'Spring' and 'Damper' Blocks generalize from 'PartialCompilant'
Create a Block 'PartialRigid' with two Ports (flanges), named 'flange_a' and
'flange_b' - these are of type Flange_a and Flange_b respectively; the 'Mass'
Block generalizes from 'PartialRigid'
Create a Block 'Fixed' with only one flange for the ceiling, which only has the
Port 'flange_a' typed to Flange_a

(c) Sparx Systems 2015 - 2017 Page 32 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

block
Oscillator
parts
fixed1 : Fixed
damper1 : Damper
mass1 : Mass
spring1 : Spring

+fixed1 +damper1 +spring1 +m ass1

block block block block

Fixed Damper Spring Mass
constraints constraints constraints constraints
{flange_a.s = s0} {v_rel=der(s_rel)} {f = c*(s_rel - s_rel0)} {v = der(s)}
{f = d * v_rel} {a = der(v)}
properties {lossPower = f * v_rel} properties {m*a = flange_a.f + flange_b.f}
s0 = 1.0 c = 10000 {flange_a.f = - m * g}
properties s_rel0 = 2
ports d = 25 properties
flange_a : Flange_a
flange_a: lossPower a
Flange_a v_rel g = 9.81
m=1
v

block block
PartialCompliant PartialRigid
constraints flange_a: constraints flange_b:
{s_rel=flange_b.s - flange_a.s} Flange_a {flange_a.s = s-L/2} Flange_b
{flange_b.f = f} {flange_b.s = s + L/2}
{flange_a.f = -f}
properties
flange_a: properties L=1
flange_b:
Flange_a f s = -0.5
Flange_b
s_rel = 0
ports
ports flange_a : Flange_a
flange_a : Flange_a flange_b : Flange_b
flange_b : Flange_b

Internal structure Create an Internal Block Diagram (IBD) for 'Oscillator'. Add properties for the
fixed ceiling, spring, damper and mass body, typed by the corresponding Blocks.
Connect the Ports with connectors.
Connect 'flange_a' of 'fixed1' to 'flange_b' of 'spring1'
Connect 'flange_b' of 'damper1' to 'flange_b' of 'spring1'
Connect 'flange_a' of 'damper1' to 'flange_a' of 'spring1'
Connect 'flange_a' of 'spring1' to 'flange_b' of 'mass1'

(c) Sparx Systems 2015 - 2017 Page 33 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

ibd [block] Oscillator [Oscillator]

fixed1: Fixed
flange_a: Flange_a

flange_b: Flange_b flange_b: Flange_b

spring1: damper1:
Spring Damper

flange_a: Flange_a flange_a: Flange_a

flange_b: Flange_b

mass1: Mass

Constraints For simplicity, we define the constraints directly in the Block elements; optionally
you can define Constraint Blocks, use constraint properties in the Blocks, and bind
their parameters to the Block's properties.

Two Oscillator Compare Plan

After we model the Oscillator, we want to do some what-if analysis. For example:
What is the difference between two oscillators with different dampers?
What if there is no damper?
What is the difference between two oscillators with different springs?
What is the difference between two oscillators with different masses?

Here are the steps for creating a comparison model:

Create a Block named 'OscillatorCompareModel'
Create two Properties for 'OscillatorCompareModel', called oscillator1 and oscillator2, and type them with the
Block Oscillator

block
OscillatorCompareModel
properties
oscillator1 : Oscillator
oscillator1: Oscillator oscillator2: Oscillator
oscillator2 : Oscillator

Setup DataSet and Run Simulation

Create a SysMLSim Configuration Artifact and assign it to this Package. Then create these data sets:
Damper: small VS big

(c) Sparx Systems 2015 - 2017 Page 34 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

provide 'oscillator1.damper1.d' with the value 10 and 'oscillator2.damper1.d' with the larger value 20
Damper: no vs yes
provide 'oscillator1.damper1.d with the value 0; ('oscillator2.damper1.d' will use the default value 25)
Spring: small vs big
provide 'oscillator1.spring1.c' with the value 6000 and 'oscillator2.spring1.c' with the larger value 12000
Mass: light vs heavy
provide 'oscillator1.mass1.m' with the value 0.5 and 'oscillator2.mass1.m' with the larger value 2

The configured page resembles this:

On the 'Simulation' page, select 'OscillatorCompareModel', plot for 'oscillator1.mass1.s' and 'oscillator2.mass1.s', then
choose one of the created datasets and run the simulation.

Tip: If there are too many properties in the plot list, you can toggle the Filter bar using the context menu on the list
header, then type in 'mass1.s' in this example.

These are the simulation results:

Damper, small vs big: the smaller damper makes the body oscillate more

(c) Sparx Systems 2015 - 2017 Page 35 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Damper, no vs yes: the oscillator never stops without a damper

Spring, small vs big: the spring with smaller 'c' will oscillate more slowly

(c) Sparx Systems 2015 - 2017 Page 36 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Mass, light vs heavy: the object with smaller mass will oscillate faster and regulate quicker

(c) Sparx Systems 2015 - 2017 Page 37 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Water Tank Pressure Regulator

In this section, we will walk through the creation of a SysML parametric model for a Water Tank Pressure Regulator
composed of two connected tanks, a source of water and two controllers, each of which monitors the water level and
controls the valve to regulate the system.
We will explain the SysML model, create it and set up the SysMLSim Configurations. We will then run the simulation
with OpenModelica.

System being modeled

This diagram depicts two tanks connected together, and a water source that fills the first tank. Each tank has a
proportionalintegral (PI) continuous controller connected to it, which regulates the level of water contained in the tanks
at a reference level. While the source fills the first tank with water the PI continuous controller regulates the outflow
from the tank depending on its actual level. Water from the first tank flows into the second tank, which the PI continuous
controller also tries to regulate. This is a natural and non domain-specific physical problem.

Create SysML Model

Component Discussion

Port Types The tank has four ports; they are typed to these three blocks:
ReadSignal - Reading the fluid level; this has a property 'val' with unit 'm'
ActSignal - The signal to the actuator for setting valve position
LiquidFlow - The liquid flow at inlets or outlets; this has a property 'lflow' with
unit 'm3/s"

(c) Sparx Systems 2015 - 2017 Page 38 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

block block block

ActSignal LiquidFlow ReadSignal

flow properties flow properties flow properties

none act : Real none lflow : Real none val : Real

Block Definition Diagram LiquidSource - The water entering the tank must come from somewhere, therefore
we have a liquid source component in the tank system, with the property flowLevel
having a unit of 'm3/s'. A Port 'qOut' is typed to 'LiquidFlow'.
Tank - The tanks are connected to controllers and liquid sources through Ports.
Each Tank has four Ports:
- qIn: for input flow
- qOut: for output flow
- tSensor: for providing fluid level measurements
- tActuator: for setting the position of the valve at the outlet of the tank
Properties:
- area (unit='m2'): area of the tank, involved in the mass balance equation
- h (unit = 'm'): water level, involved in the mass balance equation; its value
is read by the sensor
- flowGain (unit = 'm2/s'): the output flow is related to the valve position by
flowGain
- minV, maxV: Limits for output valve flow
BaseController - This Block could be super of a PI Continuous Controller and PI
Discrete Controller.
Ports:
- cIn: Input sensor level
- cOut: Control to actuator
Properties:
- Ts (unit = 's'): Time period between discrete samples (not used in this
example)
- K: Gain factor
- T (unit = 's'): Time constant of controller
- ref: reference level
- error: difference between the reference level and the actual level of water,
obtained from the sensor
- outCtr: control signal to the actuator for controlling the valve position

PIcontinuousController - generalize from BaseController

Properties:
- x: the controller state variable

block block block block

LiquidSource Tank BaseController PIcontinuousController

properties properties properties properties

cIn: ReadSignal
flowLevel : Real flowGain : Real qIn: LiquidFlow error : Real error : Real
area : Real K : Real cOut: ActSignal K : Real
ports h : Real qOut: LiquidFlow outCtr : Real outCtr : Real
qOut : LiquidFlow maxV : Real = 10 ref : Real T : Real
minV : Real = 0 tActuator: ActSignal T : Real x : Real
c onstraints
e4 :qOut: LiquidFlow
OutFlow Ts : Real
ports tSensor: ReadSignal c onstraints
qIn : LiquidFlow ports e7 : StateVariable
qOut : LiquidFlow cIn : ReadSignal e8 : OutControl
tActuator : ActSignal cOut : ActSignal
tSensor : ReadSignal
c onstraints
c onstraints e5 : CoutAct
e1 : Mass_Balance e6 : ErrorValue
e2 : SensorValue
e3 : Q_OutFlow

Constraint Blocks The flow increases sharply at time=150 to a factor of three of the previous flow
level, which creates an interesting control problem that the controller of the tank

(c) Sparx Systems 2015 - 2017 Page 39 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

has to handle.

par [block] LiquidSource [LiquidSource]

qO ut: LiquidFlow
e4 : OutFlow
{a = if time > 150 then 3*b else b}
flowLevel: Real lflow : Real
b a
equal equal

The central equation regulating the behavior of the tank is the mass balance
equation.
The output flow is related to the valve position by a 'flowGain' parameter.
The sensor simply reads the level of the tank.

par [block] Tank [Tank]

qIn: LiquidFlow
equal
e1 : Mass_Balance
lflow : Real equal {der(h) = (x - y) / a}
x y

a h
equal equal

h: Real
area: Real
e2 : SensorValue qOut: LiquidFlow
{a=b}

b a lflow : Real
equal
tSensor: ReadSignal

val : Real equal equal

e3 : Q_OutFlow
{a=LimitValue(min, max, -b*c)}
a
tActuator: ActSignal
b max min c
equal act : Real
equal equal equal

flowGain: Real maxV: Real minV: Real

The Constraints defined for 'BaseController' and 'PIcontinuousController' are

illustrated in these figures.

(c) Sparx Systems 2015 - 2017 Page 40 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

par [block] BaseController [BaseController]

cIn: ReadSignal
e6 : ErrorValue b ref: Real
{a=b-c} equal
val : Real c
equal error: Real
a
equal

cOut: ActSignal
outCtr: Real e5 : CoutAct
{a=b}
act : Real
b a
equal equal

par [block] PIcontinuousController [PIcontinuousController]

e7 : StateVariable
{der(x)=a/b} b T: Real
equal

equal a x
equal

error: Real x: Real

equal
d
e8 : OutControl
equal {a=b*(c+d)} outCtr: Real
a equal
c
b K: Real
equal

Internal Block Diagram This is the Internal Block Diagram for a system with a single tank.

ibd [block] TankPI [TankPI]

tank: Tank
source: LiquidSource
qOut: LiquidFlow qIn: LiquidFlow qOut:
LiquidFlow
tSensor: ReadSignal tActuator: ActSignal

piContinuous: PIcontinuousController
cIn: ReadSignal cOut: ActSignal

This is the Internal Block Diagram for a system with two connected tanks.

(c) Sparx Systems 2015 - 2017 Page 41 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

ibd [block] TanksConnectedPI [TanksConnectedPI]

source: tank1: Tank tank2: Tank

LiquidSourceqOut: qIn: qOut: qIn: qOut:
LiquidFlow LiquidFlow LiquidFlow LiquidFlow LiquidFlow

tSensor: tActuator: tSensor: tActuator:

ReadSignal ActSignal ReadSignal ActSignal

controller1: controller2:
cIn:
PIcontinuousController
cOut: PIcontinuousController
cIn: cOut:
ReadSignal ActSignal ReadSignal ActSignal

Run Simulation

Since TankPI and TanksConnectedPI are defined as 'SysMLSimModel', they will be filled in the combo box of 'Model'
on the 'Simulation' page.

Select TanksConnectedPI, and observe these GUI changes happening:

'Data Set' combobox: will be filled with all the data sets defined in TanksConnectedPI
'Dependencies' list: will automatically collect all the Blocks, Constraints, SimFunctions and ValueTypes directly or
indirectly referenced by TanksConnectedPI (these elements will be generated as Modelica code)
'Properties to Plot': a long list of 'leaf' variable properties (that is, they don't have properties) will be collected; you
can choose one or multiple to simulate, and they will become legends of the plot

Create Artifact and Configure

Select 'Simulate > Manage > SysMLSim Configuration Manager'

The elements in the Package will be loaded into the Configuration Manager.
Configure these Blocks and their properties as shown in this table.
Note: Properties not configured as 'SimConstant' are 'SimVariable' by default.

Block Properties

LiquidSource Configure as 'SysMLSimClass'.

Properties configuration:
flowLevel: set as 'SimConstant'

Tank Configure as 'SysMLSimClass'.

Properties configuration:
area: set as 'SimConstant'
flowGain: set as 'SimConstant'
maxV: set as 'SimConstant'
minV: set as 'SimConstant'

BaseController Configure as 'SysMLSimClass'.

(c) Sparx Systems 2015 - 2017 Page 42 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Properties configuration:
K: set as 'SimConstant'
T: set as 'SimConstant'
Ts: set as 'SimConstant'
ref: set as 'SimConstant'

PIcontinuousController Configure as 'SysMLSimClass'.

TankPI Configure as 'SysMLSimModel'.

TanksConnectedPI Configure as 'SysMLSimModel'.

Setup DataSet

Right-click on each element, select the 'Create Simulation Dataset' option, and configure the datasets as shown in this
table.

Element Dataset

LiquidSource flowLevel: 0.02

Tank h.start: 0
flowGain: 0.05
area: 0.5
maxV: 10
minV: 0

BaseController T: 10
K: 2
Ts: 0.1

PIcontinuousController No configuration needed.

By default, the specific block will use the configured values from super block's
default dataSet.

TankPI What is interesting here is that the default value could be loaded in the 'Configure
Simulation Data' dialog. For example, the values we configured as the default
dataSet on each Block element were loaded as default values for the properties of
TankPI. Click the icon on each row to expand the property's internal structures to
arbitrary depth.

(c) Sparx Systems 2015 - 2017 Page 43 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Click on the OK button and return to the Configuration Manager. Then these values
are configured:
tank.area: 1 this overrides the default value 0.5 defined in the Tank Block's
data set
piContinuous.ref: 0.25

TanksConnectedPI controller1.ref: 0.25

controller2.ref: 0.4

Simulation and Analysis 1

Select these variables and click on the Solve button. This plot should prompt:
source.qOut.lflow
tank1.qOut.lflow
tank1.h
tank2.h

(c) Sparx Systems 2015 - 2017 Page 44 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Here are the analyses of the result:

The liquid flow increases sharply at time=150, to 0.06 m3/s, a factor of three of the previous flow level (0.02 m3/s)
Tank1 regulated at height 0.25 and tank2 regulated at height 0.4 as expected (we set the parameter value through the
data set)
Both tank1 and tank2 regulated twice during the simulation; the first time regulated with the flow level 0.02 m3/s;
the second time regulated with the flow level 0.06 m3/s
Tank2 was empty before flow came out from tank1

Simulation and Analysis 2

We have set the tank's properties 'minV' and 'maxV' to values 0 and 10, respectively, in the example.

In the real world, a flow speed of 10 m3/s would require a very big valve to be installed on the tank.

What would happen if we changed the value of 'maxV' to 0.05 m3/s ?

Based on the previous model, we might make these changes:

On the existing 'DataSet_1' of TanksConnectedPI, right-click and select 'Duplicate DataSet', and re-name to
'Tank2WithLimitValveSize'
Click on the button to configure, expand 'tank2' and type '0.05' in the 'Value' column for the property 'maxV'
Select 'Tank2WithLimitValveSize' on the 'Simulation' page and plot for the properties
Click on the Solve button to execute the simulation

(c) Sparx Systems 2015 - 2017 Page 45 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Here are the analyses of the result:

Our change only applies to tank2; tank1 can regulate as before on 0.02 m3/s and 0.06 m3/s
When the source flow is 0.02 m3/s, tank2 can regulate as before
However, when the source flow increases to 0.06 m3/s, the valve is too small to let the out flow match the in flow;
the only result is that the water level of tank2 increases
It is then up to the user to fix this problem; for example, change to a larger valve, reduce the source flow or make an
extra valve

In summary, this example shows how to tune the parameter values by duplicating an existing DataSet.

(c) Sparx Systems 2015 - 2017 Page 46 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Troubleshooting OpenModelica Simulation

Common Simulation Issues

This table describes some of the common issues that can prevent a model being simulated. Check the output in the 'Build'
tab of the System Output window. The messages are dumped from the OpenModelica compiler (omc.exe), which
normally points you to the lines of the Modelica source code. This will help you pick up most of the errors.

Issue

The number of equations is less than the number of variables. You might have forgotten to set some of the
properties to 'SimConstant', which means the value doesn't change during simulation. You might have to provide the
'SimConstant' property values before the simulation is started. (Set the values through a Simulation Data Set.)

The Blocks that are typing to Ports might not contain conserved properties. If, for example, a Block 'ChargePort'
contains two parts - 'v : Voltage' and 'i: Current' - the property 'i : Current' should be defined as SimVariable with
the attribute 'isConserved' set to 'True'.

SimConstants might not have default values - they should be provided with them.

A SimVariable might not have an initial value to start with - one should be provided.

The properties might be typed by elements (Blocks or Value Type) external to the configured Package; use a
Package Import connector to fix this.

SysML Simulation Configuration Filters

The 'SysML Simulation Configuration' dialog shows all the elements in the Package by default, including Value Types,
Blocks, Constraint Blocks, Parts and Ports, Constraint Properties, Connectors, Constraints and Data Sets. For a
medium-sized model, the full list can be quite long and it can be difficult for the user to find a potential modeling error.
In the TwoTanks example, if we clear the Tank.area's property 'SimConstant' and then do a validation, we will find this
error:
Error: Too few equations, under-determined system. The model has 11 equation(s) and 13 variable(s).
This error indicates that we might have forgotten to set some of the properties to 'SimConstant'.
What we can do now is click on the second button from the right on the toolbar (Filter for the configuration) and open
the dialog shown here. Click on the All button, then deselect the 'Suppress Block' and 'Suppress Variable Part'
checkboxes and click on the OK button.

(c) Sparx Systems 2015 - 2017 Page 47 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Now we will have a much shorter list of variables, from which we can find that 'area' does not change during simulation.
Then we define this as a 'SimConstant' and provide an initial value to fix the issue.

Model Validation Examples

Variable not defined in Constraint

In the TwoTanks example, when we browse to 'constraintBlock.Outcontrol.Constraint', suppose we find a typo - we

typed 'V' instead of 'B' in the constraint.
So, instead of:
a=b*(c+d)
We typed:
a=v*(c+d)

Click on the Validate button on the toolbar. These error messages will appear in the 'Modelica' tab:
Validating model...
Error: Variable v not found in scope OutControl. (Expression: " a=v*(c+d);")
Error: Error occurred while flattening model TanksConnectedPI
Number of Errors and Warnings found: 2

Double-click on the error line; the configuration list displays with the constraint highlighted.
Change 'v' back to 'b' and click on the Validate button again. No errors should be found and the issue is fixed.
Tips: Using the SysML Simulation Configuration view is a short cut way of changing the constraints for a Block or

(c) Sparx Systems 2015 - 2017 Page 48 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

Constraint Block. You can:

Change a constraint in place
Delete using the context menu of a constraint
Add a new constraint using the context menu of a Block or Constraint Block

Duplicate Variable Names

In the TwoTanks example, browse to block.tank.constraintProperty.e1. Suppose we gave two properties the same name:
Right-click on e1, select 'Find in Project Browser', and change the name to e2; reload the 'SysML Simulation
Configuration' dialog
Click on the Validate button on the toolbar; these error messages appear in the 'Modelica' tab:
Validating model...
Error: Duplicate elements (due to inherited elements) not identical: (Expression: "SensorValue e2;")
Error: Error occurred while flattening model TanksConnectedPI
Number of Errors and Warnings found: 2

Double-click on the error line; the configuration list displays with the constraint properties highlighted.
Change the name of one of them from e2 back to e1 and click on the Validate button again; no errors should be found
and the issue is fixed.

Properties defined in Constraint Blocks not used

In the TwoTanks example, in the Project Browser, we browse to the element 'Example Model.Systems
Engineering.ModelicaExamples.TwoTanks.constraints.OutFlow'.
Suppose we add a property 'c' and potentially a new constraint, but we forget to synchronize for the instances - the
constraint properties. This will cause a Too few equations, under-determined system error if we don't run validation.
Reload the Package in the 'SysML Simulation Configuration' dialog and click on the Validate button on the toolbar.
These error messages will appear in the 'Modelica' tab:
Validating model...
Error: ConstraintProperty 'e4' is missing parameters defined in the typing ConstraintBlock 'OutFlow'. (Missing: c)
Error: Too few equations, under-determined system. The model has 11 equation(s) and 12 variable(s).
Number of Errors and Warnings found: 2

Double-click on the error line; the configuration list displays with the constraint property highlighted. The constraint
property is typed to outFlow and the new parameter 'c' is missing.
Right-click on the constraint property in the configuration list, select 'Find in All Diagrams', then right-click on the
constraint property on the diagram and select 'Structural Elements | Show Owned / Inherited' and click on 'c' and on the
Close button.
Reload the model in the 'SysML Simulation Configuration' dialog and click on the Validate button. These error messages
will appear in the 'Modelica' tab:
Validating model...
Error: ConstraintProperty 'e4' does not have any incoming or outgoing binding connectors for parameter 'c'.
Error: Too few equations, under-determined system. The model has 11 equation(s) and 12 variable(s).
Number of Errors and Warnings found: 2

(c) Sparx Systems 2015 - 2017 Page 49 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

In order to fix this issue, we can do either of two things based on the real logic:
1. If the property 'c' is necessary in the Constraint Block and a constraint is defined by using 'c', then we need to add a
property in the context of the constraint property and bind to the parameter 'c'.
2. If the property 'c' is not required, then we can click on this property in the constraint block and press Ctrl+D. (The
corresponding constraint properties will have 'c' deleted automatically.)

(c) Sparx Systems 2015 - 2017 Page 50 of 51 Created with Enterprise Architect
User Guide - Parametric Simulation using OpenModelica 30 June, 2017

(c) Sparx Systems 2015 - 2017 Page 51 of 51 Created with Enterprise Architect