International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2016): 79.57 | Impact Factor (2017): 7.296
Study of the SPH Method for Simulation in LS-
DYNA
Tran Thanh Tung
Faculty of Engineering Mechanics and Automation, VNU University of Engineering and Technology, Hanoi, Vietnam
Abstract: The aim of this paper is to presentthe capabilities of Smoothed Particle Hydrodynamics (SPH) method in LS-DYNA.A new
particle element has been added to LS-DYNA. It is based on Smoothed Particle Hydrodynamics theory. SPH is a meshless lagrangian
numerical technique used to model the fluid equations of motion. SPH has proved to be useful in certain class of problems where large
mesh distortions occur such as high velocity impact, crash simulations or compressible fluid dynamics. First, we introduce the basis
principles of the SPH method. Then the coupling of this technique to LS-DYNA is presented and the input needed for such analysis is
provided. Two types of simulations have been comducted and show once again good capabilities of this new technology for both
applications
Keywords: SPH, LS-DYNA, Simulation
1. Introduction 2. Smoothed Particle Hydrodynamics (SPH)
The Smoothed Partice Hydrodynamics (SPH) method has Smoothed Particle Hydrodynamics (SPH) is an N-body
been developed by Livermore Software Technology integration scheme developed by Lucy, Gingold and
Corporation (LSTC) in LS-DYNA since [Link] method Monaghan [1977]. The method was developed to avoid the
is called a meshfree method because traditional finite limitations of mesh tangling encountered in extreme
elements are replaced by particles which are not physically deformation problems with the finite element method. The
connected but mathematically linked [1,2]. It is an main difference between classical methods and SPH is the
alternative to the classical Lagrangian Finite Elements absence of a grid. Therefore, the particles are the
method and is used to simulate problems where materials are computational framework on which thegoverning equations
submitted to hydrodynamic deformation modes, such as high are resolved. This new model requires a new calculation
velocity impacts. This paper presents the SPH contribution method, which is briefly explained in the following.
to simulate the impact model test. Two cases to illustrate
SPH capabilities in LS-DYNA .Until now, there was a lot of 2.1. SPH formulation
research on the ability of SPH method in LS-DYNA
software. For example, Martin Madaj et al [3] presents some The particle approximation of a function is:
results of the SPH orthogonal cutting simulations of A2024-
T351 aluminium alloy compared to the experimental and
FEM simulation. Niclas Stenberg et al [4] use the SPH Where W is the kernel function i
simulation technology to predict the thermal and mechanical
loads on the rake face of aturning tool. Mohammad Sarfaraz, The Kernel function W is defined using the function θ by the
Ali Pak[5] present numerically-derived tsunami wave loads relation:
on bridge superstructures using smoothedparticle
hydrodynamics (SPH), which is a type of mesh-free
methods. D.H. Zhang [6] used SPH method with Where d is the number of space dimensions abd h is the so-
applications of oscillating wave surge converter. Jian- called smoothing length which varies in time and in space
YuChen[7] using the Smoothed Particle Hydrodynamics to W(x,h) should be centrally peaked function . The most
tackle explosion in soil with extremely large deformation common smoothing kernel used by the SPH community is
and the effects of blast on structures. RuiYan et al [8] the cubic B-spline which is defined by choosing 𝜃 as:
simulate the motion of rigid bodies bouncing from the
surface of water in two dimensions using SPH method.
Nowadays computers have become more powerful and
popular. As a result, computer simulation has become a where C is a constant of normalization that depends on the
reliable tool for many designers, engineers and scientists in on the number of space dimensions.
the investigate performance of mechanics impact models.
This study performs analysis of mechanics testing by using The SPH method is based on a quadrature formula for
SPH method in LS-DYNA software. With taking into moving particles (xi(t) ∈ 1. . 𝑁 , wherexi(t) is the location
account verification and validation (V&V) of road safety of particle i, which moves along the velocity field v.
barrier simulation, this paper presented an approach to build
a reasonable and exactly model. The results of the present The particles approximation of a function can now be
model confirmed the capabilities of Smoothed Particle defined by:
Hydrodynamics (SPH) method.
Volume 7 Issue 8, August 2018
[Link]
Licensed Under Creative Commons Attribution CC BY
Paper ID: ART2019392 DOI: 10.21275/ART2019392 271
International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2016): 79.57 | Impact Factor (2017): 7.296
In the SPH method, the location of neighboring particles is
𝑚𝑗 important. The sorting consists of finding which particles
Where wj= is the “weight” of the particles. The weight of
𝜌𝑗 interact with which others at a given time. A bucket sort is
a particle varies proportionallyto the divergence of the flow used that consists of partitioning the domain into boxes
where the sort is performed. With this partitioning the
closest neighbors will reside in the same box or in the
2.2. SPH meshing technique closest boxes. This method reduces the number of distance
calculations and therefore the CPU time.
SPH meshing technique avoids all the problems associated
with the regular Lagrangian meshing technique. The SPH The time step is determined by the expression
uses a Kernal approximation based on randomly distributed
interpolation points, with no assumptions about which points
are neighbors to calculate spatial derivatives. Where CCFL is a numerical constant
Figure 1: Calculation cycle for SPH particles
3. Implementation in LS-DYNA 3.2. Material model
3.1. Material Definitions For the SPH elements, modelled as water, air the choice is
down to two material cards. Either
SPH elements consist of only one node, on which all MAT_001_ELASTIC_FLUID can be used, which is
element properties are centered. Their initial density needs designed to model fluids, or the MAT_009_NULL. The null
to be specified and from the initial particle spacing the material can be used to model any type of homogeneous and
elements are given a certain mass. This mass stays constant isotropic fluid and solid. An equation of state needsto be
throughout the simu- lation, whereas particle density and assigned in conjunction, for which a limited number of
occupied volume may deviate from the initial value. The options are available. The Gruneisen equation of state is a
SPH processor in LS-Dyna uses a variable smoothing length good choice
[22]. The initial smoothing length h0 is computed for each
particle by taking the maximum of the distances between 3.3. Contact definitions
neighbouring particles. The smoothing length then varies in
time according to following equation: In the classic SPH technique a laborious composition of
boundary particles and ghost points is used. Within LS-Dyna
there is no need for such a construction. The software allows
mesh-based and mesh-free methods to co-exist and interact
The smoothing length decreases when particle concentration
in one simulation.
is high and increases when few particles are around. Thus, h
varies to keep the same number of particles within its
Since the SPH particles are not interconnected, only one-
support domain.
way type of contact definitions are applicable in which the
Volume 7 Issue 8, August 2018
[Link]
Licensed Under Creative Commons Attribution CC BY
Paper ID: ART2019392 DOI: 10.21275/ART2019392 272
International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2016): 79.57 | Impact Factor (2017): 7.296
SPH is always defined to be the slave and the elements are
defined to be the master. The interaction between the SPH
and other elements is defined using penalty based contact
algorithms. An appropriate con- tact type is the simple
*CONTACT_NODES_TO_SURFACE card, where the
slave nodes are checked for penetration through the master
surface. When a node is in contact with the surface, a
restoring force is applied to prevent further penetration. This
force is proportional to the penetration distance into the shell
or solid element and acts in the direction normal to the
master surface.
The restoring force is defined by:
f=kdn
In this equation, d is the penetration distance, n is the surface
normal vector and k is a penalty factor, comparable to a
spring constant. The constant k should be set large to
minimise penetration.
3.4. Simulation of impact test
Figure 2present initial condition the first simulation test. The
impact test model consist of the cylinder and the plate. The
cylinder was set 100m/s with impact angle 900as shown in
Figure 2 This model was useful method to simulate the bird
strike impact model.
Figure 3: Sequential from water drop test
Figure 4 presents the breaking waves impact with rigid
column. This case simulates the impact of a water column
on a rigid rectangle shape obstacle. The model using 192600
number of particles SPH The time analysis was set 500 ms.
The time analysis approximately15 minutes with using share
memory parallel (SMP) on one computer with 16 GB RAM,
Intel Core i7- 6700 HQ(8 CPUs).
Figure 2: The impact test simulation
The model using 126400 number of particles SPH The time
analysis was set 300 ms. The time analysis approximately3
minutes with using share memory parallel (SMP) on one
computer with 16 GB RAM, Intel Core i7- 6700 HQ(8
CPUs).
Volume 7 Issue 8, August 2018
[Link]
Licensed Under Creative Commons Attribution CC BY
Paper ID: ART2019392 DOI: 10.21275/ART2019392 273
International Journal of Science and Research (IJSR)
ISSN (Online): 2319-7064
Index Copernicus Value (2016): 79.57 | Impact Factor (2017): 7.296
[4] Niclas Stenberg,, Aldin Deli´c, Thomas Bjork. Using the
SPH method to easier predict wear in machining. 16th
CIRP Conference on Modelling of Machining
Operations. Procedia CIRP 58 ( 2017 ) 317 – 322
[5] Mohammad Sarfaraz, Ali [Link] numerical simulation
of tsunami wave forces impinged on bridge. Coastal
Engineering 121 (2017) 145–157
[6] [Link]. SPH
method with applications of oscillating wave surge
converter. Ocean Engineering .Volume 152, 15 March
2018, Pages 273-285
[7] Jian-YuChenFue-SangLien. Simulations for soil
explosion and its effects on structures using SPH method.
International Journal of Impact Engineering Volume
112, February 2018, Pages 41-51
[8] [Link]. SPH simulation of skipping
stones. European Journal of Mechanics -
B/[Link] 61, Part 1, January–February 2017,
Pages 61-71
Figure 4: Dam break simulation
4. Conclusion
This paper aims to investigate the different new features
appeared in recent versions of LS-DYNA and related with
SPH method. Different tests were done to bring some
elements in order to evaluate capabilities of these new
technologies: SPH formulations based on a Lagrangian
kernel and Hybrid SPH/Solid Elements. Results obtained
allow to be enthusiastic, first because new SPH formulations
shown good qualities concerning tensile instabilities which
stays one of the major problems with SPH nowadays. Other
limitations with this calculation method concern physical
transitions with classical Solid Elements on the one hand,
and calculation times on the other hand. That’s precisely on
that fields that new Hybrid elements are very interesting
since we saw different types of applications where this kind
of element bring obvious improvements. Large studies still
have to be done to carry on the investigation of these options
which should permit to SPH method to develop its
applications possibilities in future years.
References
[1] LS-DYNA Keyword user’s manual. Livermore Software
Technology Corporation, 2007
[2] LS-DYNA Theoretical Manual. Livermore Software
Technology Corporation, Livermore 2006
[3] Martin Madaj, Miroslav Piska. On the SPH Orthogonal
Cutting Simulation of A2024-T351 Alloy. 14th CIRP
Conference on Modeling of Machining Operations
(CIRP CMMO). Procedia CIRP 8 (2013) 152-157.
Volume 7 Issue 8, August 2018
[Link]
Licensed Under Creative Commons Attribution CC BY
Paper ID: ART2019392 DOI: 10.21275/ART2019392 274
