Costin D. Untaroiu, Jacob B. Putnam (Virginia Tech), Jeff T. Somers (Wyle Science)
New spaceflight vehicles are currently being developed to transport crews to space by NASA and several commercial companies. During the launch, and landing phase, vehicle occupants are typically exposed to spinal and frontal loading. To reduce the risk of injuries during these common impact scenarios, NASA has begun research to develop new safety standards for spaceflight. The THOR, an advanced multi-directional crash test dummy, was chosen to evaluate occupant spacecraft safety due to its improved biofidelity. Recently, a series of modifications were completed by NHTSA to improve the bio-fidelity of the THOR dummy. The updated THOR Modification Kit (THOR-k) dummy was tested at Wright-Patterson (WP) Air Base in various impact configurations, including frontal and spinal loading. A computational finite element (FE) model of the THOR was developed in LS-DYNA® software and was recently updated to match the latest dummy modifications.
Sheng Peng (LSTC)
Corpuscular particle method (CPM) was developed for airbag deployment simulations. It took into account specifics like airbag folding technique, vent hole design, and interaction between discretized gas flow and airbag fabric to capture the effects of dummy impact on airbags, both fully inflated and out-of-position. It’s numerically very robust and the particle-based nature leads to elegant treatment of venting, porous leakage and gas mixing. Users find novel situations to apply the method and oftentimes new features are needed to better support these scenarios. Among these is the need of more comprehensive treatment of heat transfer. Based on kinetic molecular theory, CPM model behavior is heavily influenced by heat transfer. Yet, a full-blown coupled thermal analysis might not always be viable for a refined model. To enable modeling of heat transfer just in the neighboring structures of the CPM gas can provide a solution to this quandary. The design and implementation status will be discussed. Some other recent advances in CPM in LS-DYNA® will also be discussed. For example, airbag integrity checking reports to the user hard-to-discover abnormality in the airbag structure definitions in the input phase.
Ioannis Tsoupis, Marion Merklein (Friedrich-Alexander-Universität Erlangen-Nürnberg)
One major challenge in metal forming exists in sheet metal bending of modern lightweight materials like high‑strength low-alloyed steels (HSLA), since conventional methods of predicting failure in numerical simulation, like the forming limit diagram (FLD), can generally not be applied to bending processes. Moreover, fracture mechanisms are mainly depending on the microstructure, which is very fine-grained in HSLA steels composed with different alloying elements compared to established mild steels. Consequently the damage and failure behaviour of HSLA steels are changing. Especially for small curvature bending processes characterised by high gradients of strain and stress over the sheet thickness other failure criteria than the FLD have to be utilised. Within this paper a numerical study of the micromechanical based damage model Gurson-Tvergaard-Needleman (GTN, *MAT_120) is performed in LS-DYNA®, in order to realise an effective adaptability of the model for bending operations on HSLA steels. The material dependent damage parameters are determined by commonly used methodology of inverse numerical identification re-calculating the uniaxial tensile test. The minimisation of the mean squared error (MSE) of experimental and numerical global load displacement curves is realised by an optimisation algorithm using commercial software LS‑OPT®. For the adaption of the GTN-Model to the bending operation a strain-based calibration method is developed. This method is based on the comparison and adaption of the numerically calculated and the experimentally measured deformation field on the outer surface of the bent specimen. In this context the parameters are systematically varied again in the optimisation software LS-OPT. Their influence on the strain and damage evolution is analysed and discussed. On the one hand it is shown that it is possible to represent the strain evolution by adapting only one parameter instead of all parameters of the model and thus reducing the modelling effort for the user. On the other hand a big effect on the damage evolution and distribution can be identified.
William Lawson, Anthony Johnson (General Atomics Electromagnetics)
To begin learning the coupled field capability of LS-DYNA and validate results, a simple simulation of parallel wires carrying current was run. The magnitude of the current in the wires is such that the coupling between the electromagnetic (EM), thermal and structural fields is weak, in the sense that the coupling is taken to be one way. That is, there is no feedback amongst the three field solutions. This allows us to compare LS-DYNA code and known analytical results for code validation to build confidence that the code is being correctly used. LS-DYNA results are also compared to ANSYS results when no analytical results are valid. In addition, this simulation allowed us to test the transfer of EM generated Ohmic heating to the thermal field, and the transfer of EM generated forces to the structural field, a necessary process for coupling fields. Furthermore, to be able to compare the code and analytical results, temperature-dependent material properties have not been included a decent approximation with the low currents used. The set-up of the coupled field model is discussed. Comparison of the LS-DYNA code and analytical results show good agreement where applicable. Comparison with ANSYS results is also good.
Kunio Takekoshi, Kazukuni Niwa (TERRABYTE Co., Ltd)
A study on preparation method of failure parameters for ductile polymers is presented using experimental results of high-speed tensile test for Polycarbonate and simulation results based on Semi-Analytical Model for Polymers (SAMP) constitutive model  in LS-DYNA®. In addition, a comparative review of two widely used failure models, namely, total formulation and incremental formulation [2, 3], is carried out using Charpy impact test simulations where the failure parameters are prepared using the proposed method. It is found that the incremental formulation is excellent in predicting the experimentally observed behavior of notched Charpy impact test and non-notched Charpy impact test.
Yih-Yih Lin (Hewlett-Packard Company)
A major hindrance to the widespread use of Implicit LS-DYNA is its high compute cost. This paper will show modern GPU, cooperating with CPU, can help remove this hindrance. Performance improvement for Implicit LS-DYNA with GPU relative to that without, as well as from recent GPU and X86 processor, will be studied. The study will cover GPU related hardware issues, including GPU Boost, memory and PCI Express Interface.
Anthony Smith (Honda R & D Americas Inc. ), Paul Du Bois (LS-DYNA® Consultant)
Generating an LS-DYNA® material model from coupon-level quasi-static experimental data, developing appropriate failure characteristics, and scaling these characteristics to mesh sizes appropriate for a variety of simulation models requires a regularization procedure. During an investigation of an anisotropic material model for extruded aluminum, numerical accuracy issues led to unrealistic mesh regularization curves and non-physical simulation behavior. Sensitivity problems due to constitutive material behavior, small mesh sizes, single precision simulations, and simulated test velocity all contributed to these accuracy issues. Detailed analysis into the sources of inaccuracy led to the conclusion that in certain cases, double precision simulations are necessary for accurate material characterization and mesh regularization.
Kirk Fraser (University of Quebec at Chicoutimi)
Smoothed Particle Hydrodynamics (SPH) has quickly become one of the most popular mesh-free methods since its introduction in 1977. In the recent years, a great amount of research has been focused on addressing some of the common computational time associated with the SPH method. One of the remaining hurdles is the long computational associated with building the neighbor list. Because of the nature of the original SPH codes (astropyshics), the neighbor search is commonly performed for every element in the domain at each time step. In this work, we develop an optimized neighbor search algorithm that is suitable for deployment on NVidia graphics cards (GPU). The SPH code is written using CUDA Fortran. The algorithm can be used for large plastic deformation computational solid mechanics (CSM) problems. The search uses an adaptive algorithm that updates the neighbor list for individual SPH elements depending whether a plastic strain increment threshold is surpassed. The neighbor list as well as the inter-particle spacing (rij) is re-used for elements that do not surpass the search update criteria. Although in this work we use a Cell based search, the algorithm can be easily adapted for the Direct Search, the Verlet List or a Tree Sort approach. Monaghan’s artificial stress term is added to the momentum equation to suppress the common tensile instability. The XSPH approach is used to update the positions of the SPH elements. The algorithm is shown to reduce the overall computation time by up to 70% without loss of accuracy for CSM simulations when compared with the non-adaptive search method.
Christian MOUTELIERE, Vincent LAPOUJADE (DynaS+), Antoine MILLECAMPS (SNECMA)
SPH, Smoothed Particle Hydrodynamics, is a very efficient tool to model industrial problems where large deformations occur. However, one disadvantage of the SPH technique is the relative expensive cpu cost compared to standard Finite Elements. Using the MPP version of LS-DYNA® allows users to handle larger problems (up to more than millions of particles) in a reasonable time. Due to the meshfree nature of the SPH method, standard decompositions used for finite elements can sometimes lead to very bad speed-up of the code. Users have to be aware of some options and rules to define customized decompositions in order to minimize communications between processors and get very good load balancing. Two classes of models are presented for addressing all possible situations with respect to optimizing MPP decomposition of a calculation based in whole or in part on the SPH technology. The first one is a pure SPH model of a high velocity impact of a sphere on a plate. The second one is a coupled FE-SPH model of a bird impacting a set of fan blades of an engine. Two versions of the same problem will be studied: for the first, shell elements are used for the modeling of fan blades whereas for the second, solid elements are used.
Dominic Schommer, Miro Duhovic, Florian Gortner, Martin Maier (Institut für Verbundwerkstoffe GmbH)
Thermoset Sheet Molding Compounds (SMC) are becoming more and more popular as lightweight construction materials in the automotive industry. SMC compression molding is a forming process in which a pre-cut SMC-Prepreg is placed within a heated mold and is first pressed into shape before being cured. By closing the mold, the thermoset resin is forced to flow and takes the randomly orientated fiber reinforcement present, along with it. The flow behavior of the SMCs can be characterized by press rheometry. In a typical press rheometry test, certain data recorded during the test, specifically press force, tool closing speed, position and time together with the known tooling geometry (plate surface area), are used to develop and verify a finite element characterization model in LS-DYNA using the relevant Arbitrary Lagrange Eulerian (ALE) capable material model. In this work, the Fluid Structure Interaction (FSI) capabilities in LS-DYNA are used to model the flow ability of the SMC material. No independent effects of the resin cure on the materials rheology are taken into consideration. The characteristic data obtained from the real press rheometry tests are used to calibrate the material model so that it can be used to predict the mold filling behavior of more complex tooling scenarios. As an example for forming a complex real part, the compression molding of a ribbed automotive spoiler test part is analyzed upon complete closing of the mold. The goal of the simulation is to provide information about the suitability of the tooling design geometry and the processing parameters. A simplified two dimensional model of the ribbed automotive spoiler part shows, that the unrecognized effects inside the material cause a failure in the simulation in certain situations. A future version of the model should make it possible to analyze the nature of the SMC part, more specifically, the flow profile, fiber orientation and resulting volume fraction in the individual sections of the part, in particular the rib section, along with chemical curing of the resin.
Li Zhang & Xinhai Zhu (LSTC)
Some of the new features developed since the last conference will be discussed. 1) Lancing – instant and progressive Cutting of sheet metal during forming to alleviate thinning and splits. 2) Auto close of open trim curve loop Improvement in trimming simulation by automatically closing an open trim curve. 3) Tailor-rolled blank thickness specification Specification of thickness field of a tailor-rolled blank for any ensuing simulation. 4) Springback compensation referencing original tool mesh Compensation using the original tool mesh for iterations to improve tool surface geometry. 5) Springback compensation – for small part shape change Compensation made easier for those parts with small shape changes that do not affect springback results. 6) Simulation-based blank size optimization A significant development in blank size and trim line optimization of stamping dies.
Xinhai Zhu & Li Zhang (LSTC)
Some of the new features developed since the last conference will be discussed. 1) Gaging pin contact improvement New contact treatment for edge contact between gaging pin and sheet blank edge during gravity loading. 2) Output control for parameterized input Specifying D3PLOT and INTFORC outputs made easy for parameterized input. 3) Gravity loading – switching between implicit dynamic and implicit static Taking advantage of the best of both dynamic and static methods. 4) Polygon adaptive box A more flexible adaptive remeshing control. 5) Maximum ID specification for blank adaptive remeshing Setting starting element and node ID for an adaptive blank in a line-die simulation. 6) Flange unfolding Unfolding of deformable flanges onto addendum for trim line development.
Akram Abu-Odeh (Texas A&M Transportation Institute)
W-beam guardrail systems are the most common roadside railing systems used by many road authorities worldwide. They have been used for decades as roadside barrier to protect errant vehicles from intruding into hazardous areas. This paper gives a description of this rail system and recent methods to simulate its performance under roadside impacts. The availability of simulation technologies such as LS-DYNA® makes it possible to evaluate the performance of guardrail systems under given impact condition. A predictive simulation example and a subsequent crash test are presented as how simulation can be integrated into roadside safety hardware design process.
C.T. Wu, W. Hu (LSTC), H.P. Wang (GM Research and Development Center)
In this paper, new numerical modeling of material flow in the thermo-mechanical friction stir welding process is presented. In this numerical model, the discretization in space is derived by the meshfree Galerkin method using a Lagrangian meshfree convex approximation. The discrete thermal and mechanical equations are weakly coupled as the time advances using a forward difference scheme. A mortar contact algorithm is employed to model the stirring effect and heat generation due to frictional contact. Heat conductance between contacting bodies is considered as a function of contact pressure. A two-way adaptive procedure is introduced to the coupled thermo-mechanical system to surpass potential numerical problems associated with the extensive material deformation and spatial discretization. In each adaptive phase, a consistent projection operation utilizing the first-order meshfree convex approximation is performed to remap the solution variables. Finally, a three-dimensional thermo-mechanical coupled friction stir welding problem is analyzed to demonstrate the effectiveness of the present formulation.
Zhidong Han, Hailong Teng and Jason Wang (LSTC)
The enhanced bond model allows the Discrete Element Method (DEM) to simulate the heterogeneity and discontinuity at the individual particle level at the micro level. The traditional material models at the macro level are applied to each particle independently. This bond model bridges the behaviors of particles at macro and micro levels, and may be used for failure analysis of the homogeneous & heterogeneous materials, including composites, concretes.
C. T. Wu, Y. Guo, W. Hu (LSTC)
This paper presents a new particle method in LS-DYNA for the severe deformation and failure analyses of solid mechanics problems. The new formulation is first established following a standard meshfree Galerkin approach for a solving of the partial differential equation of a linear elastic problem. A smoothed displacement field is introduced to the Galerkin formulation and leads to a regularized smoothed strain approximation. The resultant smoothed/regularized strain formulation can be related to the residual-based stabilization method for the elimination of zero-energy modes in the conventional particle methods. The discretized system of equations are consistently derived within the meshfree Galerkin variational framework and integrated using a direct nodal integration scheme. The linear formulation is next extended to the large deformation and failure analyses of inelastic materials. In the severe deformation range, adaptive Lagrangian or Eulerian kernel approach can be preformed to reset the reference configuration and maintain the injective deformation mapping at the particles. Several numerical benchmarks are provided to demonstrate the effectiveness and accuracy of the proposed method.
Joseph M. Dulka, Eric R. Dietrich, Kelley S. Elmore, Kendra D. Jones, Clyde S. Harmes, Robert H. Moyer (elmore engineering)
This paper presents the design process for a unique security bollard providing protection against a K12  (M50 ) vehicle crash load. LS-DYNA was used to aid in the design of the security bollard to account for the highly dynamic and inelastic behavior during a vehicle impact. The bollard was installed along the top of a wall for a below-grade courtyard in order to maintain a building security perimeter, providing protection against a potential malevolent vehicle attack. Contrary to typical bollard installations where the foundation is supported on all sides with well compacted soil or other substrata, no significant support was provided on the protected side of the bollard foundation. As a result, this posed significant difficulty in the design of an effective security bollard required to resist a potential K12 (M50) vehicle impact load with zero vehicle penetration. The initial conceptual bollard design originated from the standard Department of State (DoS) DS-22 K-12 rated bollard system . Hand calculations were used to develop a preliminary bollard design with equivalent static design stopping forces based upon existing physical K12 test results. LS-DYNA aided the engineering team in observing structural and material responses characteristic of impact loading which may have otherwise not been perceived by method of traditional hand calculations. Utilizing LS-DYNA as not only an analysis tool, but a powerful design tool enabled the engineering team to optimize the design of the security bollard.
Yusuke Nakae, Tsuyoshi Yasuki, Hiroshi Tanaka, Jiro Takamitsu (Toyota Motor Corporation)
This paper describes the numerical analysis of unsteady aerodynamics of a car model in dynamic pitching motion using LS-DYNA R7. Large-Eddy simulations with ALE method were performed to clarify the effects of unsteady aerodynamic forces on aerodynamic characteristics of cars in dynamic motion. The forced sinusoidal pitching oscillation was imposed on the 1/4 scaled car model and the flow velocity was set to 27.78 m/s. The model was based on a real production car and it was simplified by removing its engine compartment cavity and smoothing its surface. Tires were fixed on the ground and separated from the car body. Unsteady aerodynamic forces acting on the model were investigated. And the mechanism of the differences between the aerodynamic forces acting on the car model in the dynamic motion and those in stationary states occur was mainly discussed. The computational results showed good agreement with the results of the high accuracy LES code computations. Also, results showed the differences between the aerodynamic forces in the dynamic pitching motion and those in the stationary states. Especially, the lift force showed remarkable differences. Even in the same posture of the pitch angle 0 degree (i.e. the posture in which the under floor of the car body is parallel to the ground), the lifts showed different values at stationary state and during nose-up or nose-down respectively. As a result of this analysis, it was revealed that these differences in the aerodynamic forces were mostly due to the changes of the surface pressure distributions around rear end of the front wheelhouse. The flow structures behind the front tires changed with volume shrinking or expanding of front wheel house owing to the car motion. These changes affected the surface pressure distributions.
Helga Kuhlmann, Pieter Volgers, Zhenyu Zhang (DuPont)
In the field of structural design, from aerospace and automotive to consumer packaging, numerical structural analysis using the finite element method (FEM) is becoming ever more important to accurately predict the performance of the considered part. In a highly competitive market, industries are demanding higher performance, improvements in fuel efficiency, increased recycling and greater safety, whether this is an airplane wing or mineral water bottle. In response to the above factors, there has been a significant increase in the application of composites. Today, finite element simulations are used extensively in the design and assessment by virtually all mayor industries. Finite element analysis (FEA) has become an integrated tool in this design and optimization. In this paper, beams constructed from over molded Short fiber Reinforced Thermoplastic on Continuous Fiber Reinforced Thermoplastics are described. One of the challenges is accurate CAE simulation of the static and dynamic behavior of the part. Model data are validated through correlation between coupon and sub-system physical tests, and further verified with results from quasi-static and impact tests. Physical test on beams confirmed good correlation between test and Finite Element Analysis.
Yun Huang, Zhe Cui (LSTC)
NVH (Noise, Vibration and Harshness) is an important topic for the design and research of automotives. Increasing demands for improved NVH performance in automotives have motivated the development of frequency domain vibration and acoustic solvers in LS-DYNA. This paper presents a brief introduction of the recently developed frequency domain vibration and acoustic solvers in LS-DYNA, and the application of these solvers in auto NVH problems. Some examples are given to illustrate the applications.
D. Marzougui, C.D. Kan, K.S. Opiela (George Mason University)
The introduction of new crash test requirements raises questions about the efficacy of commonly used barriers that had been accepted under earlier test requirements. Seven commonly-used barriers were crash tested under the new MASH requirements in a recent NCHRP project. Three of the barriers tested did not meet the new requirements for the test with the 2270 kg vehicle. While the implementation of the MASH standards does not require hardware that passed the previous NCHRP 350 requirements to be re-evaluated, there is an interest in knowing whether these devices can be modified to meet the more stringent MASH requirements by DOTs. In another effort, these seven NCHRP crash tests were successfully simulated to provide an extended validation of the new finite element model of a Chevrolet Silverado pick-up truck as a surrogate for the 2270 kg test vehicle. This provided the opportunity to, among other things, evaluate the potential of various retrofit options for improving two of the three barriers that failed. An analyses of six modifications for the G9 Thrie-beam barrier and three variations of the G4(1S) guardrail median barrier was undertaken. A summary of the testing and simulation modeling of the two tests is presented as the basis for the simulation of modified versions of the barriers. The evaluation results are presented for each of the retrofit options and recommendations offered.
Yun Huang, Zhe Cui (LSTC)
This paper presents the new ATV (Acoustic Transfer Vector) and MATV (Modal Acoustic Transfer Vector) techniques for BEM acoustics in LS-DYNA, which were implemented recently. Acoustic Transfer Vector provides the transfer function between the normal nodal velocity on structural surface and the acoustic response at selected field points; Modal Acoustic Transfer Vector provides similar transfer function, but is based on the excitation from modal shape vibrations. ATV and MATV reveal the inherent properties of structures and acoustic volume, and can be used to predict radiated noise from vibrating structures when combined with vibration boundary conditions. Particularly they are useful for the acoustic analysis of structures subjected to multiple load cases. Some examples are given to illustrate the application of the ATV and MATV techniques. For ATV, post-processing of the results in the form of binary plot database is also presented.
Ryan Alberson, David Stevens (Protection Engineering Consultants), James D. Walker, Tom Moore (Southwest Research Institute)
Software has been developed to automatically mesh CAD files in support of expedient modeling of armored vehicles and similar structures. The AutoMesher software is written in Python as well as LSTC’s Script Command Language (SCL). The SCL syntax is similar to C programming, but runs as a script within LSTC’s LS-PrePost® (LSPP) software application. A Python module is used as the interface and a wrapper for LSPP. By leveraging the functions in LSPP through the SCL, nine different algorithms were written to mesh I beams, T beams, angles, rods, plates, tubes, and surface-meshed formed shapes. Logic is used in these algorithms to identify the shape characteristics needed to define an equivalent FEA mesh of the CAD geometry, such as geometric planes that represent flanges or web components of an I-beam. These algorithms are the heart of the AutoMesher and can be used to generate more intelligent meshing solutions. The algorithms and software are described in this presentation. The AutoMesher software was developed by Protection Engineering Consultants (PEC) in support of the Defense Advanced Research Projects Agency (DARPA) Adaptive Vehicle Make (AVM) program, under subcontract to Southwest Research Institute (SwRI). AVM is an ambitious program to reduce the time required for the design, development, and production of complex defense cyber-mechanical systems, such as military ground vehicles, by a factor of five.
Jennifer Yee (Combat, University of Massachusetts), James A. Sherwood (University of Massachusetts), Stephen Fitzgerald (Combat)
The ideal models of softball bats and balls should have the flexibility to allow for the ability to capture how BBS varies as a result of changes in bat and ball constructions. If such models were available, then the design engineer could customize the bat design with the goal to maximize the BBS for a given ball construction. A credible finite element model of the ball-bat collision for softball is challenging. Achieving such a model is difficult primarily because of variations in the processing of the polyurethane cores of softballs which can yield different properties of the overall ball, e.g. hardness and liveliness, and the response of the softball during a bat-ball collision is rate dependent. The mechanical behavior of the composite bat is slightly less challenging to model because the bat material can be assumed to be essentially linear elastic unless significant material damage is induced during the collision. Experimental and finite element methods were used to model the collision between a composite softball bat and softballs of different COR (Coefficient of Restitution) and compression specifications. An example model is shown in Figure 1. Experimental bat characterization methods included barrel compression and modal analysis. Experimental softball characterization methods included COR, CCOR (Cylindrical Coefficient of Restitution), compression and dynamic stiffness. Finite element models were built in HyperMesh and analyzed in LS-DYNA®. Softballs were modeled using LS-DYNA material models #6, #57 and #83, and the composite softball bat was constructed according to the manufacturer’s specifications using *PART_COMPOSITE. Three methods to calibrate the finite element softball models were investigated and included “flat-surface” and “cylindrical-surface” coefficients of restitution and DMA (dynamic mechanical analysis). The “cylindrical-surface” test was found to be the most effective method of calibration to predict the batted-ball speed (BBS) as measured in bat/ball impact testing. This paper presents a summary of the complimentary experimental and finite element studies that were completed to develop a bat-ball collision model for the research of composite softball bats. Softballs were characterized using simple tests, and finite element models of the softballs were calibrated to yield good correlation to the experimental characterization tests. The calibrated softball models were then used to explore their ability to correlate with bat-ball collisions using a composite softball bat.
Myeong-Gyu Bae, Seonwoo Lee (THEME Engineering, Inc.)
The composite material is widely used in the structures as aircrafts, satellites, ships, automobiles, and so on which demand light weight and high performance. A various type of damage could occur through low-speed impacts and fatigue loads. It is generally known that the assessment of the natural frequency by vibration testing is a very attractive method as a Non-Destructive Test (NDT) and the vibration response of a composite structure can be utilized as an indicator of damage. In this paper, a desirable FE modeling technique regarding composite types (Laminated/Sandwich) and laminate methods of composite material were investigated using LS-DYNA. Firstly, according to the laminated properties of composite material (number of layers, anisotropy, shape, etc.), frequency responses were compared between the latest theories and the latest version of LS-DYNA. Secondly, various types of damage in cantilever beam with composite material were represented and estimated in FE model and those frequency responses were compared among experiments, LS-DYNA, and other FE code. Finally, delamination phenomenon in rectangular plate with composite material was represented and estimated in FE model and those frequency responses were compared between experiment and LS-DYNA. It was evaluated and verified that the prediction for the tendency of natural frequency using the frequency domain method in LS-DYNA could be appropriate for composite materials with or without damage.
Marcus Lilja (DYNAmore Nordic AB)
The use of non-linear FEA is growing in the offshore industry. Det Norske Veritas AS (DNV), one of the world’s largest ship and offshore classification societies to the maritime industry, has developed a Recommended Practice (DNV-RP-C208) on the usage of non-linear implicit finite element simulations in offshore applications. DNV-RP-C208 creates a de facto standard for structural load capacity analysis of off-shore structures. The Recommended Practice in combination with handbook formulas and empirical data create a de facto standard that will be used for investigations, studies and dimensioning off-shore structures for years to come. The Recommended Practice contains several benchmark problems with references solutions that can be used to verify a finite element software and the modeling methodology. This paper presents results from LS‑DYNA for a selection of these benchmark problems , ranging from beam bending problems with elasto-plastic behavior to instability and collapse analysis. All benchmark problems are solved using the implicit non-linear solver. Development of new features in LS‑DYNA and LS-PrePost® were necessary in order to complete the task. This paper presents results from the benchmarks, solution techniques, and the newly developed features.
Eric L. Ruggerio, James A. Sherwood and Patrick J. Drane (University of Massachusetts)
The bats used in Major League Baseball (MLB) are turned from a single piece of wood. Northern white ash had been the wood of choice until the introduction of hard rock maple in the late 1990s. Since the introduction of maple, there has been a measurable increase in the number of bats breaking into multiple pieces. These failures can be a significant factor during play, i.e. pieces of bats landing flying into the field of play, thereby distracting fielders from making the appropriate play for the given game situation. Observations of bat breakage in the field and in controlled conditions of lab testing of bats have shown the bat durability is a function of wood quality and bat profile. Wood quality is described by the density and the slope of grain of the wood. The bat profile is described by the variation in the diameter of the bat along its length. The bat properties that are preferred by players, i.e. low-density wood and a bat profile of a big barrel and a slender handle, are in direct contradiction with what makes for a durable bat. In this paper, LS-DYNA is used to develop calibrated models of the breaking of yellow birch wood bats in controlled lab conditions. The WOOD material model in combination with the ADD EROSION option using a maximum principal strain failure criterion was found to produce a credible simulation of the failure modes seen in wood baseball bats.
Bill Feng (Jaguar Land Rover)
Curtain airbag is a key restraint component to protect occupants in the events of side impact (referred as First Impact) and rollover (referred as Second Impact). In the curtain airbag design during the vehicle programme, following requirements dominate the design. FMVSS226 Ejection Mitigation (EjM) requires curtain airbag provide adequate protection for rollover event. Restraint system performance for legal and consumer tests, such as FMVSS214 and NCAPs, requires good occupant head protection in the first impact. TWG Out-Of-Position (OOP) requires low risk deployment of curtain airbag for the occupants seating in out-of-position. In addition, curtain airbag design should ensure the integrity of surrounding trims, such as pillar trims, during the deployment at different environmental conditions. In 2011, NHTSA introduced the new regulation for rollover protection, FMVSS226 Ejection Mitigation. The requirement demands increased occupant containment in rollover and side crashes for belted/unbelted occupants and third rows of seating. The rule requires the linear impact tests at two energy levels and two inflation times (e.g. 278J@1.5s and 178J@6s). The results of this requirement are the introduction of larger curtain airbag with higher power inflator for longer inflation. Since then, FMVSS226 EjM has become a key loadcase to define the curtain airbag inflator selection and curtain airbag design. However, the introduction of larger curtain and larger inflator has great challenge to the integrity of curtain airbag and surrounding trims, and OOP performance as well. Therefore, it is important that balanced performances between restraint system requirements and component requirements during the process of curtain airbag design and inflator selection. In this paper, CAE applications and studies have been conducted to gain the understanding of energy requirements and managements for balanced curtain design airbag to meet the multiple requirements on restrain system performance, EjM, OOP and component integrity.
Henry Shibayama, Rohit Ramanna, Sri Rama Murty Arepalli, Arthur Camanho (ESI)
Visual-Environment is an open and integrated user environment enabling simulation and analysis across multiple domains for efficient product engineering. A CAE workflow has been developed chaining stamping and impact simulations. The workflow originates from stamping matrix design (performing stamping simulations) to impact simulations considering the residual stress and thickness variation due to stamping process. The objective to be achieved is the creation of a fast end-to-end workflow, aiming at accurate impact simulations while taking into account the results from manufacturing processes. So far, impact simulations are performed considering the stamping simulation results. The next step of the project is to perform welding simulations and consider its residual stresses and distortion, aiming at more accurate impact simulations, through chaining and considering the process effects coming from stamping and welding analysis as well. Visual-Diemaker is a software tool focusing on the design of the stamping matrix, with some feasibility tools, such as tipping evaluations. Visual-Diemaker also integrated some tools for the set-up of stamping simulation. For development and evaluation of the methodology all simulations were performed using PAM-STAMP. Generated output results were M01 files (one per component), containing the residual stress and thickness variation. In a first step Visual-Process, a mass-customization and automation tool to support the automation of CAE tasks, converts automatically M01 (PAM-STAMP) file structure to LS-DYNA® key-words respectively syntax (using ELEMENT_SHELL_THICKNESS and INITIAL_STRESS_SHELL). As a second step, the same tools allows the CAE engineer to open the LS-DYNA impact model in Visual-Environment, then setting the components to be chained with the stamping results, define the reference nodes for the construction of a reorientation matrix, and finally define a different number of integration points through the thickness. This process basically reads and converts the M01 syntax and also adds the INCLUDE_STAMPED_PART syntax into the global impact model which references the converted M01 file. The purpose of the project is the ability of automated chaining between manufacturing and performance structural simulation in one and same environment. After achieving this development goal, further implementation and industrialization of this kind of analysis methodology into a CAE industrial department is expected as a reliable and fast way to proceed with chained analysis using different CAE solver.
Vito Primavera, Marco Perillo (EnginSoft SpA), A. Carofalo, M. De Giorgi, R. Nobile (University of Salento)
Metallic foams are very promising for engineeristic applications due to their peculiar characteristics, like the high energy-absorbing property coupled with a reduced weight. Even if applications can be widespread in several fields, such as automotive, civil, aerospace, etc., industrial requirements are still far to be fully accomplished, especially in terms of technological processes and a whole mechanical characterization. Material modeling of metallic foams, like the aluminium ones, is a crucial point for performing accurate numerical simulations along with the design phase. Material models available in the explicit, non-linear finite element code LS-DYNA® represent a very efficient way to handle and to investigate foam behavior. An extended experimental/numerical activity has been set out at the aim to calibrate and validate suitable material models with respect to different aluminium foams and several loading conditions. While a previous phase of the activity  has been focused on the assessment of a procedure addressed to point out, starting from the available experimental data, the key points of material model calibration, the current activity has been focused on the procedure application, i.e. the exploitation of the built-up methodology in respect of calibration of M-PORE open cells aluminium foam at three different loading conditions. A good number of foams material models are available in the LS-DYNA database, and further in the last years different enhancements have been performed at the goal to include the physical phenomenons able to increase the accuracy of the models. Amongst the available ones, MAT 154 (MAT_DESHPANDE_FLECK_FOAM) has been here chosen because it provides satisfactory results compared with the experimental ones, but at the same time it still requires to be studied for more loading conditions. Since the calibration process requires to optimize the material model free parameters according to different objectives, LS-DYNA has been coupled with modeFRONTIER®, Process Integration and Design Optimization software platform. Once all the FE (Finite Element) models related to the corresponding experimental tests have been integrated into modeFRONTIER, a first sensitivity analysis has been performed at the purpose to get confidence with MAT 154 behavior and then an efficient optimization phase in order to pursue the numerical configurations satisfying the different targets provided by experimental tests. Efficient and intuitive post-processing tools have been applied firstly to get a deep knowledge of the investigated phenomenons and eventually to look for the best solutions.
Phani Adduri, Gary Quinn, Juan P. Leiva, Brian C. Watson (Vanderplaats Research and Development, Inc)
This paper demonstrates a design system to efficiently perform optimization based on responses computed from multiple LS-DYNA® analyses while also taking into consideration the linear loading conditions such as the ones for NVH and Static responses. The proposed design system, ESLDYNA, is based on the Equivalent Static Load (ESL) method, which requires the iterative process of non-linear structural analysis (LS-DYNA) and linear structural analysis and optimization (GENESIS). Unlike general purpose optimization software packages, it does not require a large number of analysis calls even for problems with large numbers of design parameters. Therefore, large-scale optimization techniques, such as topology, topometry and topography, can be easily employed. Several examples using different optimization techniques will be presented. One of the examples will include optimizing the design for frontal crash, normal modes and static loading conditions simultaneously.
Teddy MAILLOT, Vincent LAPOUJADE, Edith GRIPPON (DynaS+), Bernard TOSON, Nathalie BARDON, Jean-Jacques PESQUE (CEA CESTA)
In order to compute the requirements for transporting packages in various situations using balsa wood as an energy absorber, a constitutive model is needed that takes into account all of the specific characteristics of the wood, such as its anisotropy, compressibility, softening, densification, and strain rate dependence. Completeness alone is not sufficient for the model, because it must perform appropriately in simulations that include many other non-linear situations, such as being subjected to friction, undergoing large deformations, and even failure. To improve their existing modeling within LS-DYNA, CEA CESTA, in partnership with I2M of Talence, carried out a major experimental campaign both on standard characterization tests and on more complex tests representative of the behavior of real structures. All these tests have been modeled using different LS-DYNA material laws to assess their respective limitations and achieve optimal modeling within the framework of material laws currently available in LS-DYNA. In a final validation phase, this optimized material law has been introduced in a finite element model representative of a real package to evaluate its effect relative to the initial law.
Sunao Tokura (Tokura Simulation Research Corporation)
SPH (Smoothed Particle Hydrodynamics) implemented in LS-DYNA® has been used widely in various industrial fields as a reliable and robust particle method. At present SPH is considered as one of major numerical simulation method for compressible fluid and solid materials. Recently a unique particle method called MPS (Moving Particle Simulation) has been developed and started to use for some industrial application as a CFD (Computational Fluid Dynamics) solver for incompressible flow. As most application for fluid flow in industry are incompressible, MPS may have a potential ability to treat such problems efficiently than SPH. Both methods have common characteristics that particles are used to discretize continuum domain to be solved. However, as the numerical procedures to solve the governing equation are very different, each numerical simulation method has both inherent advantages and disadvantages. This paper demonstrates the comparison of SPH and MPS for some engineering problems and intends to reveal the difference of these two methods. Comparison of numerical simulation techniques should be very useful for further understanding about multiphysics capability of LS-DYNA even for expert LS-DYNA users. Surface tension model, turbulence model, treatment of Newtonian and Non-Newtonian fluid, coupling with structures and other several topics are discussed. In addition an FSI (Fluid Structure Interaction) problem using MPS software and LS-DYNA is demonstrated in the presentation. In this FSI problem a vehicle is washed away by a tsunami and crashes against a rigid wall. Pressure of tsunami on the surface of the vehicle is computed by MPS software and the deformation of the auto body is calculated by LS-DYNA.
Rahul Makwana (DEP-Autoline Inc), Sumit Sharma (Eicher Trucks and Buses VE Commercial Vehicles Ltd.), Liying Zhang (Wayne State University)
Impact induced traumatic brain injury (TBI) has been studied by physical testing using various surrogates, including cadavers, animals, and crash test dummies and by computer modeling including Finite Element (FE) models of human, animal and crash test dummy head. The blast induced TBI research and evaluation of a protective device call for a head model which can mimic wave propagation phenomena through different parts of the head. For proper investigation of head responses and resulting brain injuries due to primary blast exposure, the characteristics of a physical test headform including details of brain/skull anatomy and material properties of the head tissues must be critically designed. The current study was undertaken to numerically evaluate the blast performance of an anatomically realistic headform constructed with existing skull/brain simulant materials in comparison with human head model responses in order to propose a future headform which could be used for testing equipment in blast loading conditions. Quantitative biomechanical response parameters such as pressure, strain and strain rates within the brain were systematically monitored and compared between the blast anatomical headform and the FE human head model. The results revealed that the blast anatomical headform resulted in an average of about 20% over prediction of the biomechanical response parameters in the brain. The results imply that the plyometric based thermoplastic, polycarbonate, polymethylmethacrylate, and polyoxymethylene can be the suitable surrogate skull materials for simulating head responses under blast exposure.
Uli Göhner, Bruno Boll (DNYAmore GmbH), Inaki Caldichouri (LSTC), Tim Wicke (Volkswagen AG)
Due to the increasing demands on lightweight design, stiffness and crash performance of automotive body components, the press hardening method becomes widely-used. The high strength of press hardened parts of up to 1.5 GPa results from the nearly complete conversion of austenite into martensite. This microstructural transformation, also known as ‘hardening’, happens during or subsequently to the forming process. In order to achieve a cooling rate which is high enough to get a martensitic microstructure in all regions of the blank, it has to be ensured that the heat transfer rate from the blank to the tool and inside the tool is sufficiently high. This is achieved at the press hardening lines of Volkswagen through the cooling of the tools with a fluid.
Pierre L’Eplattenier, Julie Anton, Iñaki Çaldichoury (LSTC)
The Electromagnetics (EM) solver of LS-DYNA® has recently been extended to shell elements, in order to solve coupled EM/mechanical/thermal problems on thin plates, which appear in Magnetic Metal Forming and Welding experiments. Due to the magnetic diffusion of the EM fields through the thickness of the plate, which is a very important phenomenon that needs to be precisely solved, the EM part of the simulation still needs a solid mesh with several through thickness elements. This solid mesh, underlying the shell mesh is thus automatically built during the simulation and is used to solve the EM equations. The EM fields are then averaged or summed through the thickness in order to compute equivalent EM fields on the shells, and in particular an equivalent Lorentz force and Joule Heating which are used by the mechanical and thermal solvers. The model is presented and illustrated on some academic and industrial examples. Comparisons between solid and shells are presented.
Ulrich Stelzmann, Madhukar Chatiri (CADFEM GmbH), Thorsten Schütz (Adam Opel AG), Anton Matzenmiller (Univ. of Kassel)
The objective of this paper is to present a workflow for numerical modeling and simulation of carbon fiber reinforced plastic (CFRP) composite structures including CAE process integration. A computational constitutive model for anisotropic damage is developed to characterize the elastic-brittle behavior of fiber-reinforced laminated composites. The composite damage model is implemented within LS-DYNA® as user defined material subroutine. A CAE process chain which includes the manufacturing side of composites is also presented.
Katharina Fischer (KTM Technologies GmbH), Phelippe Pereira (ESSS), Madhukar Chatiri, Matthias Hörmann, Andre Stühmeyer (CADFEM GmbH)
The goal of this presentation is to study the structural behavior of the KTM “X-BOW” crash box front impact structure in a 0° impact test against a rigid wall. The energy absorbing crash box is made of laminated composite sandwich material. A “shell-solid-shell” numerical approach is used to model the sandwich composite structure. Shell elements are used for the face layers whereas solid elements are used for aluminum honeycomb core. Shell elements consider the composite layering using *ELEMENT_SHELL_OFFSET_COMPOSITE within LS-DYNA® and will be bonded to the solid elements without node sharing. The composite structure is modeled using *MAT_054 and honeycomb structure is modeled using *MAT_126 within LS-DYNA. For comparison reasons, numerical and experimental results for intrusion, deceleration, velocity and displacement over time are presented.
D. Marzougui, C.D. Kan, and K.S. Opiela (George Mason University)
Detailed finite element (FE) models of a 2270 kg Chevrolet Silverado and a 1100 kg Toyota Yaris are used as surrogates for barrier crashworthiness under the new Manual for Assessment of Safety Hardware (MASH). MASH requires assessment of barriers for both large and small vehicles, hence the use of 2270P and 1100P test vehicles. Impacts of these two vehicles into a New Jersey-shaped concrete median barrier were simulated and compared to full-scale crash tests. The objectives of this effort included (1) demonstrating the viability of the FE models for the new MASH crashworthiness evaluation, and (2) describing the application of the newly developed roadside verification and validation (V&V) procedures to compare simulation results and crash test data. Comparisons of the simulation results and data derived from crash tests using “traditional” methods suggested that the models provided viable results. Further comparisons using the new V&V procedures provided (1) a structured assessment across multiple factors reflected in PIRT tables and (2) statistical comparisons of the test and simulation results allowing a more robust validation than previous approaches. These comparisons further confirmed that the new vehicle models were able to effectively replicate impacts for MASH tests and that the V&V procedures provided useful insights and increased confidence in the models.
Thomas Borrvall (DYNAmore Nordic AB), Dilip Bhalsod, John O. Hallquist, Brian Wainscott (LSTC)
Subcycling in explicit finite element simulations refers to the technique where a model is partitioned in levels of the characteristic time step of its constituting finite elements. Each sub model is then integrated independently of the others using a time step that pertains to that specific sub model, with the exception of special treatment at the interface between sub models. With the subcycling option in LS-DYNA, up to seven sub models are automatically generated, each integrated in steps of 1, 2, 4, 8, 16, 32 and 64 times the smallest characteristic time step of the entire model. To allow more control of the partition, the user may manually designate parts to be integrated at specific time steps. This is sometimes referred to as multiscale since it is mainly intended for detailed modeling of critical components in a large simulation model, i.e., different time scales are used in order to save CPU time. This paper presents the current status of this feature in LS-DYNA, including a detailed description of the involved algorithms and presentation of small to large scale numerical examples.
Gilles Marchaud, Louis Vilela, Stéphane Nallet (AREVA TN)
For 50 years, AREVA TN has been supplying customer-focused, innovative transportation and storage solutions for radioactive material with the highest levels of safety and security. Transportation and storage casks are designed to comply with stringent regulations. For instance, the TN NOVA™ system, designed to store used fuel assemblies, is required to withstand the impact of a 20-ton aircraft at a velocity of 215m/s, despite the extremely small probability of such an event actually occurring. The TN NOVA™ system is composed of a sealed NUHOMS®-69BTH Dry Shielded Canister and a TN NOVA™ Overpack. The overpack has been designed to house the canister during the storage period and provide it with an efficient protection against airplane crash events. To achieve this, LS-DYNA® was invaluable in helping us to improve the preliminary design and to select the most damaging airplane impact configuration. LS-DYNA® analyses also made it possible to design an equivalent missile that causes deformations at least equal to those caused by an airplane crash. The equivalent missile model was updated thanks to a real test onto a concrete wall. Finally, the overpack design was successfully validated by a real test. The equivalent missile impacted a 1/3 scale mock-up of the canister-loaded overpack, fitted with strain gages and accelerometers. Leak tightness was preserved. The present paper will focus on the crashworthiness LS-DYNA® calculations and benchmarks that made this success possible.
William W. Feng, John O. Hallquist (LSTC)
In this paper, the strain-energy density with Mullins damage function on unloading and subsequent reloading is considered. We introduce a damage function that has four material constants: two for unloading and two for subsequent reloading. The effect of these constants on unloading and subsequent reloading is studied for uniaxial extension. We determine these four material constants from a set of numerically generated uniaxial extension test data. The mathematical formulation has been implemented in LS-DYNA® for user application and evaluation. This paper will be extended to two-dimensional problems and a set of biaxial test data will be obtained and analyzed. The second part of this paper will be presented in another LS-DYNA conference.
D. Marzougui, D. Brown, H.K. Park, C.D. Kan, and K.S. Opiela (George Mason University)
A Finite Element (FE) model of a mid-size passenger sedan was created by reverse engineering to represent that aspect of the fleet in crash simulation analyses. A detailed FE model of this vehicle was created to allow application for different types of crash scenarios. The initial version of the model includes detailed and functional representation of suspension and steering components. Material characteristics and thicknesses of the different components were determined from manufacturer’s information and coupon tests so that the simulated crash behavior would reflect actual impact test results. The model mass and inertial properties were compared to measurements made on the actual vehicle. Initially, the model was subjected to a series of debugging and verification simulations to insure that all components of the vehicle are included and appropriately connected. A series of validation tests followed to compare simulated and actual crash tests. Comparisons to full-scale crash tests indicated that acceleration pulses at different locations of the vehicle, deformations in the occupant compartment, and overall vehicle kinematics are similar. Detailed representation of the vehicle interior components and restraint systems is currently being incorporated in the model to provide opportunities to use FE occupant models in the vehicle and assess injury risks.
Amos Gilat, Jeremy D. Seidt, Jeremiah T. Hammer, and Timothy J. Liutkus (Ohio State University, USA)
Calibration and verification of simulations with LS-DYNA® in which plasticity and failure models, like MAT224, are used require data from well controlled experiments. One example is the simulation of containment during blade-off and disk failure in jet engines. This application requires accurate simulation of the penetration of a projectile at many combinations of impact speeds, projectile-target geometries, and temperatures. To validate the simulations of this application, a new dynamic punch test has been developed. In this test, shown schematically in the figure below, a round disk specimen is attached to the transmitter bar of a compression Split Hopkinson Bar (SHB) apparatus, and a punch is attached to the incident bar of the SHB apparatus. During a test, a compression wave is introduced into the incident bar which causes the punch to penetrate into the specimen. The full-field deformation of the back surface of the specimen is measured during the test by using the Digital Image Correlation (DIC) technique. This is possible because the specimen is supported by a slotted tubular adaptor that provides a stereographic view of the deforming back surface. The force measured in the transmitter bar of the SHB apparatus corresponds to the contact force between the punch and the specimen. Various states of stresses and different penetration modes (petaling, bending, plugging) can be obtained by changing the specimen thickness and punch geometry. Results from tests with specimens made of Ti-6Al-4V with punches of various geometries show that the punch geometry greatly influences the punching force and the failure mode. The 3D DIC and the force measurements provide data that can be used to construct and validate deformation failure models.
Niclas Brannberg, Pere Fonts, Chenhui Huang, Andy Piper, Roger Malkusson (Qoros Automotive Co., Ltd)
Pedestrian protection has become an important part of the Euro NCAP consumer test and to achieve a 5-star rating for crash safety, a good rating for the different pedestrian load cases is imperative. It was decided at the very start of the Qoros 3 Sedan’s development program that this should be made a priority. A skilled team of safety engineers defined the layout of the vehicle to support this target, and an extensive simulation program using LS-DYNA was planned to define and validate the design intent without compromising the design of the vehicle as well as maintaining all other important vehicle functions. This paper will provide an insight into part of the journey taken to establish a new vehicle brand in China, fulfilling high European safety standards at an affordable cost, and how Qoros succeeded in this mission with a combination of skills, extensive CAE analysis and finally the validation of the recipe during physical testing. The paper will highlight how the high rating for pedestrian protection was obtained and give a short overview of the complete safety development of the Qoros 3 Sedan.
Nobuhisa Ishiyama, Shinobu Tanaka, Satoshi Fukushima, Tsuyoshi Yasuki (Toyota Motor Corporation), Masahiro Saito (Toyota Technical Development Corporation), Jesse Buehler, Brian Tew (Toyota Motor Engineering & Manufacturing North America)
This paper describes a finite element model for a Researched Moving Deformable Barrier (RMDB) that simulates an oblique crash test. National Highway Traffic Safety (NHTSA) is currently conducting research on oblique RMDB-to-Vehicle (Oblique) testing. The RMDB, which consists of an aluminum honeycomb and an outer cladding sheet, exhibited two deformation features after the oblique crash test. The first was cracks observed on the outer cladding sheet. The second was compressive deformation, mainly observed on the 0.724MPa aluminum honeycomb. The RMDB FE model was developed based on the SAE paper. The aluminum honeycomb had two layers with different stiffness and was modeled by shell element to capture compressive deformation. The outer cladding sheet was modeled by tied overlapping shell elements, in order to simulate crack propagation. The RMDB FE model was validated through the impactor test and the full car test. The results of the analyses using the model closely matched to the test results. The impactor model was developed to conduct impactor component testing. The aluminum honeycomb was glued to the jig and the impactor crashed into the aluminum honeycomb. The resulting fracture line on the outer cladding sheet and impactor acceleration data was correlated to test. Next, full car testing was performed refer to the SAE paper. The RMDB and car kinematics, velocity, structure deformation, and body intrusions largely matched to those from test. Cracks, generally corresponding to those in the test, were observed from analysis result in the outer cladding sheet. The aluminum honeycomb compressive deformation was also close to the test deformation result. Investigation of the effects of crack propagation in the outer cladding sheet revealed that deformation in the upper aluminum honeycomb showed difference depending on whether the outer cladding sheet had cracks ore not. Thus, reproducing the outer cladding sheet cracks is effective in simulating RMDB deformation.
Ming Cheng, Doug Bueley, Lock-Sui Chin, Jean-Philippe Dionne, Neil Wright, Aris Makris (Med-Eng Holdings, LLC)
In addition to traditional threats to vehicle occupants from frontal crash and side impact, passengers of military vehicles can also be subjected to vertical shock loading on their thoraco-lumbar spine and legs arising from the detonation of roadside bombs or landmines. In such explosion events, the vehicle hull is subjected to high level transient momentum loading, resulting in an acceleration impulse that transfers to the occupant through the vehicle floor and the seat. When conducting experimental blast testing, full-scale anthropomorphic test devices (ATD) are used to evaluate the survivability potential of passengers. Equivalent investigations involving ATD models are also conducted numerically. However, existing ATD numerical models have been developed mainly for frontal crash and side impact simulations, and have not been validated against vertical impact loading experienced by military vehicle occupants. The purpose of this paper is thus to compare blast loading results obtained with several ATD numerical models in a representative scenario. As a baseline, a simple drop test was experimentally conducted with a 50th percentile male ATD sitting on a platform, to simulate the vertical impact from a blast. Through the use of this simple structure, uncertainties arising from complicated seat and test fixture structures were avoided. During the test, the assembly consisting of the ATD and platform was placed in a controlled drop tower facility to generate an impact pulse on the ATD. The pelvis acceleration, lower lumbar force, and neck force were recorded. Simulations of this test were then performed with LS-DYNA® using different numerical ATD models. It was found that none of the numerical ATD models investigated could generate accurate enough responses, when compared to the experimental test with the physical ATD model. To extend observations from the above comparisons to more practical loading scenarios, blast off test simulations were also conducted and the results were compared with the signals recorded in an experimental blast off test. It is thus concluded that further enhancements to numerical ATD models are required for simulating occupant responses under vertical loading.
Chin-Hsu Lin, Yi-Pen Cheng (General Motors), Jason Wang (LSTC)
A uniform pressure method, i.e. no pressure variation on bag surface and location, in LS-DYNA has been commonly used to simulate airbag deployment and interaction of airbag with the occupants. Another newly developed LS-DYNA CPM (Corpuscular Particle Methodology) has gained recognition and acceptance recently because it considers the effect of transient gas dynamics and thermodynamics by using a particle to represent a set of air or gas molecules and then a set of particles to represent the entire air or gas molecule in the space of interest. This innovative method, however, has yet be fully utilized and applied with confidence in airbag deployments simulation without systematic tests and validations to avoid non-physical tuning factors traditionally being applied to the uniform pressure airbag finite element models. In this paper, inflator closed and vented tank tests, static airbag deployment test, and linear impactor tests with various configurations and impact speeds are systemically conducted and then correlated with a CPM airbag model to determine whether the methodology can be applied for all the tests and whether any tuning factors should be applied in the process. This innovative LS-DYNA particle method has been fully investigated in this systematic study by correlating it with a comprehensive set of inflator tank tests, static airbag deployment, and rigid linear impactor tests. The correlations start from inflator closed and vented tank tests to verify the provided inflator characteristics, mass flow rate and temperature curves. The inflator characteristics will then be employed into static airbag deployment simulation to determine the airbag fabric heat convection coefficient, which is adjusted in this simulation to match the test pressure profile. This is the only parameter tuned to match the test pressure. This airbag model is then used to simulate those linear impact tests. With the systematic validations and correlations to avoid using tuning factors, the airbag model results in a good match of the overall airbag internal pressure and impactor deceleration histories with the tests and the simulations for all the linear impactor tests conducted. Effects of the inflator variations are also studied to illustrate the potential bounds of deceleration and airbag chamber pressure in impacts.
G. Huang, H. Zhu, S. Sriram (ArcelorMittal Global R&D E. Chicago Center), Y. Chen, Z. C. Xia, O. Faruque (Ford Motor Company)
Reliable predictions of the fracture behavior in a crash event have become ever important in recent years as they will enable the reduction of physical prototype testing and the acceleration of vehicle development time while maintaining high safety standards. The increasing use of even stronger grades of Advanced High-Strength Steels (AHSS) such as hot-stamped boron steels provides particular challenges to fracture modeling due to their microstructures and processing conditions. This paper provides a brief description of the different fracture criteria and their implementation currently available in LS-DYNA to model ductile failure. The focus is the determination of parameters for selected fracture criteria for AlSi coated press -hardenable steels using calibration tests at the coupon level and supported by FEA simulations.
M. Duhovic, M. Hümbert, P. Mitschang, M. Maier (Institut für Verbundwerkstoffe GmbH), P. L’Eplattenier , I. Çaldichoury (LSTC)
Continuous induction welding is an advanced material processing method with a very high potential of providing a flexible, fast and energy efficient means of joining together thermoplastic composites to themselves and metal alloys. However, optimization of the process is very difficult as it involves the interaction of up to four different types of physics. In the previous installments of this work, static plate heating and continuous induction welding simulations of carbon fiber reinforce thermoplastic (CFRTP) plates were presented looking in particular at point temperature measurements and 3D surface plots of the in-plane temperature distribution across the entire width of the joint on the top as well as the joining interface of the laminate stack. In this paper, static plate heating tests are once again revisited and the importance of through the thickness temperature behavior is considered. For a single plate, the through thickness temperature profile follows a predictable pattern when using an induction frequency producing a skin depth of the same thickness as the plate. For two stacked but unconnected plates, the temperature profile becomes less obvious, in particular for plate stacks of different thicknesses. By correctly simulating the through thickness temperature profile the heating behavior can be ultimately controlled via top surface air-jet cooling together with other induction equipment parameters giving an optimum heating effect at the joining interface. In addition, further developments in the induction heating electromagnetism module available in LS-DYNA® R7 are examined including the inclusion of an orthotropic electromagnetic material model as well as electrical contact and its resulting contact resistance and effect on the overall heating behavior.
Brian Walker, Liliana Cowlam, Jamie Dennis (Arup), Simon Albery, Nicolas Leblanc (Futuris)
It is essential for seat manufactures to be able to accurately predict the H-Point position of a seat during the design stage, i.e. before the seat is actually built. This can be estimated empirically but this method is usually not sufficient to accurately determine how the manikin’s position is affected by subtle yet complex interactions within the seat and its trim. To aid this process, Arup have developed a positioning tool kit for use in conjunction with the Oasys PRIMER software . The positioning tool kit calculates the H-Points of the automotive seats as well as the backset measurement thus providing the scores of the head restraint. The benefit to the seat engineer of using the Oasys HPM positioning tool is increased confidence in the H-Point of a new seat design, and an opportunity to adjust the design to minimise H-point variation that may be measured in test. This improved understanding of the seat will allow more accurate predictions of whiplash performance and other crash test simulations where dummy positioning is critical.
Jorgen Bergstrom, David Anderson, David Quinn, Eric Schmitt, Stuart Brown, Samuel Chow (Veryst Engineering, LLC)
The increased use of polymeric materials in impact and high strain rate applications is motivating the use of impact simulations during design. However, simulation of polymer impacts requires difficult-to-measure stress-strain behavior at high strain rates. Even when appropriate data is collected, accurate high strain rate constitutive models need to be fit to the data before being incorporated into a simulation code. This article presents a testing and constitutive modeling process using Veryst’s PolyUMod® and MCalibration® to achieve accurate impact simulations using polyether ether ketone (PEEK) as the example material. Low and high strain rate data is presented over a large strain rate range. Validation of the developed material model is performed by simulating a drop test in LS-DYNA® with comparison to measured drop test data.
Danghe Shi, Xinran Xiao (Michigan State University)
A large amount of work has been done to simulate the crashworthiness of composite structures, particularly to evaluate the deformation behavior and to determine the energy absorbing efficiency. However, the existing simulation models generally need to introduce many non-measurable parameters which limited their practical applications. This work focused on the implementation and development of a thermodynamically consistent continuum damage mechanics (CDM) model called Ladevèze model. This model took into account stiffness recovery and inelastic strains, both damage and plastic strains. All the parameters needed in this model can be determined by experiment. Modified Ladevèze models were developed in order to adapt different damage and plasticity evolution laws for different fabric forms of composites. Three different versions of Ladevèze model were implemented in LS-DYNA and their predictive abilities were studied.
Shota Yamada (Fujitsu Limited)
In modern high performance computing era, parallel computing has been a trend to improve the speed of computation. In the past we have found that just simply increasing the number of computing parallelism would not guarantee to achieve better performance especially when simulating large deformation using hundreds or more number of parallel processors. Through our past experience, to improve the computational performance, we had found it was necessary to tackle on the issue of load unbalance of calculation cost among processors and to seek for better strategy in domain decomposition. In general, calculation cost increases with respect to the extent of deformation. To reduce the unbalance of calculation cost among processors, ideally we would like to decompose domain to subdomains with same extent of deformation on all processors. Even it is possible, it would be difficult to achieve such ideal decomposition for the cases with only local deformation occurred in crash simulation. Therefore we come up with a new enhanced method to decompose the model by distributing calculation cost more uniformly in crash simulation. In this paper, I will reveal this enhanced method, present the results of improved performance of this method using several models of crash simulation, and discuss the efficiency of this method.
Adrian Jensen, George Laird (Predictive Engineering), Kirk Fraser (Predictive Engineering, University of Quebec at Chicoutimi)
The Discrete Element Method (DEM) is fast becoming the numerical method of choice for modelling the flow of granular material. Mining, agriculture and food handling industries, among many others, have been turning their attention towards this powerful analysis technique. In this paper, we present three simple calibration modeling tactics that should be the starting point for every DEM simulation of dry and semi-dry granular material. The three tests are designed to be as simple as possible in order to minimize the run time of the test simulations. The tests are developed to be run in a specific order, providing a sequential calibration procedure that does not involve multiple unknown variables in each test. Other standard testing methods are briefly discussed, such as the rotating drum and the shear cell (Jenike) tests. The complexity of these tests does not lend itself well to initial numerical model calibration as each test involves many unknown variables. However, they are mentioned as an extension of the three basic test models. The paper will help analysts to increase the precision and validity of their discrete element modelling work.
Olivier Schreiber, Tony DeVarco, Scott Shaw, Aaron Altman (SGI)
SGI delivers an unified compute, storage and remote visualization solution to our manufacturing customers that reduces overall system management requirements and costs . LSTC has now integrated Explicit, Implicit solver technologies into a single hybrid code base allowing seamless switching from large time steps transient dynamics to linear statics and normal modes analysis. There are multiple computer architectures available from SGI to run LS-DYNA. They can all run LSTC solvers using Shared Memory Parallelism (SMP), Distributed Memory Parallelism (DMP) and their combination (Hybrid Mode) as supported by LS-DYNA. Because computer resources requirements are different for Explicit and Implicit solvers, this paper will study how advanced SGI computer systems, ranging from multi-node Distributed Memory Processor clusters to Shared Memory Processor servers address the computer resources used and what tradeoffs are involved. This paper will also outline the SGI hardware and software components for running LS-PrePost® via SGI VizServer with NICE Software. CAE engineers, at the departmental level, can now allow multiple remote users create, collaborate, test, optimize, and verify new complex LS-DYNA simulations in a single system and without moving their data.
Subramani Sockalingam, Michael Keefe, John W. Gillespie Jr. (University of Delaware)
High performance polymer fibers such as Kevlar, Spectra and Dyneema are widely used in ballistic impact applications. Under transverse compression at finite strains these fibers exhibit nonlinear inelastic behavior. The role of transverse compression during ballistic impact is not very well understood. In this work we implement a transversely isotropic inelastic constitutive model as a user defined material model (UMAT) in LS-DYNA®. A plasticity approach is used to model the material nonlinearity and a pseudo-elastic approach for the large residual strains in the transverse fiber plane. Based on the experimental results, the material nonlinearity and inelasticity are decoupled from the fiber direction. The UMAT predictions for a single Kevlar KM2 fiber under transverse compression are compared to the experimental load deflection under monotonic and cyclic loading.
Jingxiao Xu, Jason Wang (LSTC)
Smooth particles hydrodynamics is a meshfree, Lagrangian particle method for modeling fluid flows and solid bodies. It has been applied extensively to the multiphase flows, heat conduction, high explosive problems and so on. In this paper, different interaction methods available in the LS-DYNA for SPH parts which have wide range of density and material properties are studied and compared. Node to node contacts fit well for the interaction between two SPH parts with high density ratio, the standard SPH interpolation method has better accuracy around the interfaces when two SPH parts have similar density and material properties. Different interaction approaches can be combined together in one model to reach the best results. Also the interactions between Lagrangian elements with SPH particles are discussed. Some examples are presented to show how to use different approaches with different combination of LS-DYNA keywords.
Chunjie Zhang, Philip Ho, Xinhai Zhu (LSTC)
Die system module (DSM) is developed to generate tool geometry in an early stage and to evaluate these result by forming simulation. DSM Graphics User Interface is designed to provide metal forming users a tool to generate die face more effectively. The main focus of module is placed on easy modification and reuse of existing design. This paper illustrates the algorithm and some special feature of DSM
Liping Li, Roger Grimes (LSTC)
Rotor dynamics is commonly used to analyze the behavior of structures ranging from jet engines and steam turbines to auto engines and computer disk drives. In such applications, the amplitude of structural vibration can become excessive when the speed of rotation approaches the system’s critical speed. This paper introduces a primary implementation of rotor dynamics in LS-DYNA and presents a validation study of this new implemented feature with exiting theoretical studies, as well as another finite element method software ANSYS. The structural vibration responses of four different models with beam, shell and solid elements, the shaft whirling orbit and Campbell diagrams are compared. It shows that the results from LS-DYNA have very good agreements with theoretical results and ANSYS simulation results. So it suggests that the LS-DYNA simulation is accurate for the cases investigated in this paper.
S.A. Muflahi, G. Mohamed, S.R. Hallett (University of Bristol)
Predictive capabilities to simulate the initiation and propagation of delamination in thin composite laminates have been investigated. Different element formulations (3D solids, 2D shells, and 3D thick shells), cohesive fracture models (commercially available in LS-DYNA 971 v6.1 and *USER_DEFINED constitutive behavior) and stacking procedures have been applied to representative composite models of increasing complexity to demonstrate their response, delamination failure modes and computational efficiency. It has been shown that stacks of 2D shell elements with nodal offsets with a user-defined constitutive model for cohesive elements can retain many of the necessary predictive attributes of delamination dominated failure while providing superior computational efficiency and flexibility required for industrial component scale design.
Attila P. Nagy, David J. Benson (Dept. of Structural Engineering, UCSD), Stefan Hartmann (DYNAmore GmbH)
Two new areas of development of isogeometric analysis in LS-DYNA are presented. The first, which is currently available, is mass scaling. The second, which will be available sometime during the next year, is the development of efficient integration methods for trimmed NURBS, which will allow a much more direct connection between CAD and analysis in LS-DYNA. Industrial applications of both are presented. Metal stamping is one of the most cost effective manufacturing methods for producing precision parts. Isogeometric analysis, which uses the same basis functions as the CAD programs used to design the shape of the part, is an attractive alternative to traditional finite element analysis for metal stamping. Mass scaling, and the underlying stable time step estimates, that are commonly used in metal stamping simulations are presented for isogeometric analysis. Additionally, a numerical algorithm is proposed to construct efficient quadrature rules for trimmed isogeometric elements as part of the standard pre-processing step. The motivation is to overcome the proliferation of quadrature points observed in competing adaptive and tessellation-based integration approaches. The constructed integration rule is considered to be optimal in the sense that the final quadrature points and weights satisfy the moment fitting equations with the trimmed domain up to a predefined tolerance. The resulting quadrature points are in the interior of the trimmed domain and positivity of the weights is preserved. The efficiency and accuracy of the scheme is assessed and compared to competing integration techniques. Selected problems of elastostatics and elasto-plastic dynamics are used to further demonstrate he validity of the approach.
Noriyo Ichinose (JSOL Corporation)
Recently vehicle modelling is becoming more detailed and complex. Automotive companies are more and more directly evaluating dummy injury criteria in crash analysis. To evaluate injury criteria, a more detailed seat model is needed, because injury criteria are highly depending on seat structure and restraint system. In addition to the above, many types of LS-DYNA analysis are carried out during one seat design process (e.g. frontal impact, side impact, whiplash, and so on). Because these analyses use different dummy models, different loading conditions and sometimes different dummy/seat positions, the engineer needs to understand all regulations and make a big effort to prepare the input data. To reduce this effort in the demand for more detailed seat models, an integrated seat design system named JSD has been developed.
Anil Kalra, Feng Zhu, King H Yang, Albert I King (Wayne State University)
A numerical simulation is conducted to model the explosive detonation and blast wave propagation in the open air field. The mesh size and boundary conditions as well as size of air domain are the sensitive variables which may significantly affect the predicted pressure wave magnitude and rising time in blast simulations. The current approach focuses on determining the optimal key parameters to predict the blast wave accurately. A 2D to 3D mapping is performed to save the computational time. The blast induced high pressure waves are generated using the Arbitrary Lagrangian-Eulerian (ALE) formulation in the 2D domain and then mapped into a 3D space. The simulation results show that the aforementioned parameters govern pressure wave form in both 2D and 3D cases. A two-step mesh sensitivity study is performed: A parametric study is first conducted in the 2D air domain and then followed by a second one in the 3D domain while using 2D to 3D mapping. After that, as a case study in the biomedical applications, an anatomically detailed pig head finite element model is integrated with the 3D air domain to calculate the pressure gradient change inside the brain due to blast wave. The model predictions are compared with the experimental data and it has shown that the modeling strategy used can capture the biomechanical response of the surrogate with reasonable accuracy and reduced computational cost.
Hao Chen, Jason Wang, Ian Do (LSTC)
LS-DYNA ALE(Arbitrary Lagrange-Eulerian Method), equipped with its own fluid-structure interaction, aims to solve a series of transient engineering problems characterized by large momentum and energy transfer between Lagrange structures and ALE fluids. LS-DYNA ALE multi-material formulation solves multiple species of fluids in one ALE mesh. The fluid interfaces are tracked internally by our interface reconstruction algorithms at each of the advection cycle. Then our fluid-structure interaction algorithm is used to study the interactions between structures and those individual fluids. The FSI solver, invoked by the *CONSTRAINED_LAGRANGE_IN_SOLID card, is to couple between ALE fluid elements and Lagrange structure segments. The multi-material capability, together with its embedded coupling to structures, have been utilized by users from various engineering application areas such as tank sloshing, tire hydroplaning, bottle dropping, high explosive blasting, etc. Several recent developments and their engineering applications in LS-DYNA ALE/FSI package are presented here.
Suri Bala (LSTC)
LS-DYNA Data Processing, Storage and Visualization consume a lot of time and effort for every Engineer and Scientist who uses simulations to aid product development. This paper reviews commonly used workflows in simulation based product design to identify areas where d3VIEW can significantly reduce time and effort in data intensive tasks. In conclusion, this paper will demonstrate by example how d3VIEW provides advanced capabilities in data extraction, organization and visualization of LS-DYNA simulations to expedite the process of going from Data to Decision while providing extensive capabilities in mining historical LS-DYNA simulations.
Gilbert Queitzsch (Federal Aviation Administration), Cing-Dao Kan, Kivanc Sengoz (George Mason University), Thomas J. Vasko (Central Connecticut State University)
In addition to the well-known parallel versions of LS-DYNA, the symmetric multiprocessing (SMP) version and massively parallel processor (MPP) version, LSTC offers an LS-DYNA HYBRID version that combines these two parallel programming models into a single code. The development of LS-DYNA HYBRID, which started in 2011, is focused on obtaining high code performance on large cluster environments. The intent of the current study is to investigate the LS-DYNA HYBRID performance, scalability, and output consistency using a modified LS-DYNA Aerospace Working Group Generic Fan Rig Model. The original model is an outcome of a Federal Aviation Administration (FAA) funded university project and it is used as a test case for the LS-DYNA Aerospace Working Group Test Case Suite.
Alexander Akkerman, Yijung Chen, Bahij El-Fadl, Omar Faruque, Dennis Lam (Ford Motor Company)
LS-DYNA has been used for vehicle crash simulations for many years. The models have increased in size over the years but in most cases do not exceed more than a few million elements. However, recently developed material models require much greater levels of refinement resulting in much larger models, perhaps as high as 100M elements. Simulating models of the order of 100M elements in turn requires much higher levels of scalability in order to be feasible in the vehicle development process. This paper will analyze LS-DYNA performance with a 100M-element sled side impact model running on up to 1,000 and more CPUs with various Intel processors and Infiniband interconnect technologies.
Facundo Del Pin, Iñaki Çaldichoury, Rodrigo R. Paz (LSTC)
LS-DYNA R7 introduced an incompressible flow solver which can track flow interfaces such as free surfaces or the interface between two fluids. Several industrial applications may be simulated with these features. In the area of free surface flows the effects of the lighter phase are neglected, i.e. in the case of water-air interfaces the air could be ignored if its effect does not change significantly the dynamics of the water phase. Some typical problems are wave propagation, dam break, sloshing problems and green water on decks. On the other hand problems where both phases should be taken into account are mixing problems, bubble dynamics and lubrication problems among others. In this work examples of both problems will be presented and explained. The set up process as well as the post processing will be detailed. Validation examples will be shown and compared to analytical or experimental solutions. Finally the current development status for some of the multiphase features will be discussed.
Iñaki Çaldichoury, Facundo Del Pin, Rodrigo R. Paz (LSTC)
LS-DYNA version R7 introduced an incompressible flow solver (ICFD solver) which may run as a standalone CFD solver for pure thermal fluid problems or it can be strongly coupled using a monolithically approach with the LS-DYNA solid thermal solver in order to solve complex conjugate heat transfer problems. Some validation results for conjugate heat transfer analyses have been presented at the 9th European LS-DYNA Conference (2013) . This paper will focus on a new output quantity, the heat transfer coefficient or ‘h’ which has recently been implemented in the ICFD solver. Its description, calculation and uses will be presented as well as some validation results.
Ting-Ting Zhu (Cray Inc.), Jason Wang (LSTC)
For the automotive industry, car crash analysis by finite elements is crucial to shortening the design cycle and reducing costs. To increase the accuracy of analysis, in additional to the improvement in finite element technique, smaller cells of finite element meshes are used to better represent the car geometry. The use of finer mesh coupled with the need for fast turnaround has put increased demand on scalability of the finite element analysis. In this paper, we will use the car2car model to measure LS-DYNA scalability on Cray® XC30™ supercomputers, an Intel ® Xeon® processor-based system using the Cray Aries network. The scalability of different functions in LS-DYNA at high core counts will be analyzed. The MPI communication pattern of LS-DYNA will also be studied. In addition to that, we will also explore the performance difference between using one thread per core and two threads per core. Finally, we will explore the performance impact of using large Linux Huge Pages.
John Hallquist, Yun Huang, Iñaki Çaldichoury, Jason Wang (LSTC)
Recent enhancements – John Hallquist Linear solver – Yun Huang LS-PrePost: ICFD & EM – Iñaki Çaldichoury Particle methods – Jason Wang
Nielen Stander and Anirban Basudhar (LSTC)
New features available in LS-OPT® 5.1 are discussed and illustrated. The main features include three new solver types, Parallel Feedforward Neural Networks, seamless variable de-activation for iterative methods, exporting of selected metamodel formulae, subregion-based Global Sensitivity Analysis, enhanced histogram visualization features and Viewer-based categorization of simulation results.
Trevor Dutton, Paul Richardson (Dutton Simulation Ltd)
A key part of the build-up to the London 2012 Olympic Games was the Torch relay for which each one of the 8,000 runners required a Torch. The design of the Torch comprised inner and outer skins of perforated aluminium formed into a triangular cross-section, which flared out towards the top to house the gas burner. Dutton Simulation was asked to assist with development of a process to manufacture the skins to the required accuracy and quality of finish; some of the key technical challenges are described in the paper. The first task was to develop a blank shape for the two forms and then confirm these with incremental forming simulation (using eta/DYNAFORM with the LS-DYNA® solver). The validated shapes – both the profile and the thousands of holes – were then cut by laser. In conjunction with developing the blank, the optimum forming process also had to be determined, to form the perforated sheet to the accuracy required for laser welding the joining seam. Several process concepts were explored before arriving at a four stage method. With aluminium as the raw material springback was already expected to be a factor; this was compounded by the holes further reducing the material stiffness and the relatively low strain in the form due to the large radii. Nonetheless, the geometry had to be formed to a very tight tolerance, both for the weld process and also to create a result free of cosmetic defects. LS-DYNA was used to determine the springback at each step of the forming process (Figure 1) and the springback compensation solution was used to provide the correction. DYNAFORM’s tools for cosmetic defect detection (stoning, reflect lines) were employed to check the result to the highest level of detail.
Pak Lui, Gilad Shainer, Scot Schultz (Mellanox Technologies, Inc.)
From concept to engineering and from design to test and manufacturing, the automotive industry relies on powerful virtual development solutions. Crash simulations are performed in an effort to secure quality, safety and accelerate the development process. As the models become more complex to better simulate the physical behavior in crash simulations, the computers that run as a cluster also need to be higher to meet the needs of the higher standards for simulating these more elaborate models. Among the various components in a compute cluster, the high performance network interconnect is an integral factor which is key in making the simulation run efficiently. The Mellanox Connect-IB™ InfiniBand adapter has introduced a novel high-performance and scalable architecture for high-performance clusters. The architecture was designed from the ground up to provide high performance and maximize scalability for the largest supercomputers in the world today and in the future. This paper demonstrates the new features and technologies driven by the Connect-IB InfiniBand adapters. Besides its raw abilities of delivering sub-microsecond latency and a full bandwidth of over 100Gbps using two of the FDR links, its hardware capabilities also includes CPU offloads, MPI collective operations acceleration and message transport services that make LS-DYNA to perform at scale. This paper also demonstrates running multiple parallel simulations to achieve higher cluster productivity, in an effort to exploit with this new level of performance available from the network.
Yijung Chen, Omar Faruque, Cedric Xia, Alex Akkerman, Dennis Lam (Ford Motor Company)
The scope of this paper focuses on the characterization and prediction of potential crack initiation and propagation in a boron-steel component under extreme impact load, utilizing a meso-scale FE (0.2 mm solid element) modeling with the MIT MMC (modified Mohr-Coulomb) fracture criterion. The MMC fracture criterion is implemented through LS-DYNA® *MAT224 and *MAT_ADD_EROSION with GISSMO option. A finite element mesh with total number of elements close to 100 million is created to investigate the accuracy of MMC criterion in predicting fracture of a boron component in a dynamic impact test. The CAE results are compared to sled test results for system force-deflection, part deformation mode and crack initiation and propagation.
R. Reichert, C.-D. Kan, D. Marzougui, U. Mahadevaiah, R. Morgan, C.-K. Park, F. Tahan (George Mason University)
Integrated occupant-vehicle analysis plays an important role in vehicle and occupant safety developments. Car manufacturers are using detailed full system models consisting of vehicle structure, interior, restraint systems, barrier, and occupant to develop safety measures and assure compliance with legal requirements, good rating results in consumer information tests, and vehicle safety in real life crash configurations. Suppliers are using sub-system models to design and optimize interior and restraint system components with respect to various component and system requirements. This paper describes efficient methodologies for fully integrated occupant-vehicle simulations as well as sub-system evaluations using prescribed motion in LS-DYNA®. Examples include different short duration impacts such as frontal and side impact configurations with termination times of less than 200 milliseconds, and long duration impacts such as rollover events with termination times of 400 to 2500 milliseconds. A frontal offset and a frontal oblique impact was simulated using a Toyota Yaris model, side impact simulations were conducted with a Ford Taurus model, and a Ford Explorer model was used for rollover evaluations. Occupant models used include a Hybrid III, a THOR (Test device for Human Occupant Restraint), a US side impact, and a WorldSID dummy, as well as a THUMS (Total HUman Model for Safety) human model. Simulation results are compared to available full-scale crash test data. Parametric studies have been conducted to examine the influence of different input and output parameters when using sub-models with prescribed motion.
Derek Nevins, Lloyd Smith (Washington State University)
Finite element modeling of dynamic sports ball impacts presents a substantial challenge. This is because, rather than displaying linear-elastic behavior, many sports balls are predominantly non-linear, inelastic and rate dependent. This is true of both softballs and baseballs, which exhibit strong rate-dependence and large energy dissipation characteristics in collisions occurring under play-like conditions. The development of finite element models of these balls is further complicated by the difficulty in measuring materials properties at strain rates and magnitudes representative of play. This work describes the development of novel ball models from data obtained under play-like conditions. Ball models were implemented in LS-DYNA® using the Low-Density Foam material model. Simulations were compared to empirical data collected over a range of ball speeds. Models displayed good agreement with experimental measures of energy dissipation and impact force and represent an improvement over commonly used viscoelastic models.
M.S. Hamid (Advanced Computational Systems, LLC), Minoo Shah (IDIADA Automotive Technology)
Concussion, as known as mild Traumatic Brain Injury (mTBI), is the most common sport-related head injury. Football is the most common sport with higher concussions in USA. Helmet is the equipment being used in mitigation of mTBI. There are numerous designs of helmets which meet the requirements of sport regulation committee. In this paper, a football helmet is evaluated using numerical methods. The brain and the tissues in human head are modelled using continuum Smoothed Particle Hydrodynamics (SPH). The brain tissues are generated by segmentation from human brain MRI data. The LSTC dummy is used to represent the football players. The brain tissue is fitted in the cavity of the dummy headform. Two different impact scenarios are simulated in this study. The results for these impact conditions are presented.
Seung Hun Jeong (10DR KOREA Co., Ltd.)
This paper will focus on the main features, benefits and use of MME-Converter and MME-Report which could be highly useful to LS-DYNA users for vehicle crash test analysis. With MME-Converter users can simply convert LS-DYNA result files, such as nodout, elout, deforc and rcforc to MME-filtered files through the auto-syntax analysis and then compare these converted MME-filtered files to real vehicle crash test data. The conversion of LS-DYNA files is carried out in accordance with the international occupant protection criteria including KNCAP, USNCAP, Euro NCAP and IIHS. Furthermore, MME-Report which is one-page reporting system using MME-filtered data helps users to create concise, professional engineering reports, so that engineers in CAE teams, could share the test results with each other and even use them for formal meetings or presentations.
Roger Grimes (LSTC)
In LS971 R71, LSTC has enhanced its capabilities in Modal Dynamics from previous versions. This talk will give an overview of the enhanced capabilities which include mode selection and modal damping. We will present an industrial example of using this capability including a comparison of using Modal Dynamics and a full simulation.
W. Zhao, J. Liu, W. Stilwell, B. Hempy, Z. Karoutas (Westinghouse Electric Company LLC)
As a primary barrier to the fission product release, maintaining the structural integrity of fuel rod cladding has been a topic of great importance. To help better understand the structural behavior of the fuel rod in shipping and handling incidents, a detailed model for a typical pressure water reactor (PWR) fuel rod is being developed using LS-DYNA. The paper describes an on-going model development effort. For efficiency of the development process, a shortened version of the fuel rod is considered with full fuel pellet stack represented by five pellets. Nevertheless, the model contains all the structural features of the fuel rod, thus can be easily extended to obtain a full length fuel rod model.
Kyoung-Su Im, Zeng-Chan Zhang, and Grant Cook, Jr. (LSTC)
Airbags are part of an important vehicle safety system, and the inflator is an essential part that generates a specific volume of gas to the airbag for a short duration of time. Recently, we have developed numerical models of automotive airbag inflators in conjunction with the LS-DYNA® chemistry solver. In this presentation, we will demonstrate two different models: a conventional pyrotechnic inflator and a compressed, heated gas inflator. Detailed and comprehensive descriptions for constructing the keyword flies will be given and the results for the two models will be discussed. Limitations of the currently available models and future directions for coupling with the existing LS-DYNA® solvers, i. e., ALE and CESE solvers will also be presented. In addition, more advanced models will be proposed and discussed in detail.
Leonard E Schwer (Schwer Engineering & Consulting Services)
As part of the “Blind Blast Simulation Contest 1 ,” organized by the University of Missouri Kansas City, participants were invited to submit predictions of reinforced concrete slabs subjected to air blast loading. There were two classes of concrete: normal strength f c ′ = 5 ksi (34.5 MPa) and high strength f c ′ = 15 ksi (103.5 MPa). The normal strength concrete was reinforced with Number 3 Grade 60 steel bars with yield strength of 68 ksi (469 MPa). The high strength concrete was reinforced with Vanadium Number 3 bars with nominal yield strength of about 83 ksi (572 MPa). Each concrete slab design was subjected to two different air blast wave forms with impulses of about 5.38 and 7.04 MPa-ms. For the purposes of this reinforcement modeling study, the normal strength f c ′ = 5 ksi (34.5 MPa) concrete reinforced with Number 3 Grade 60 steel bars with yield strength of 68 ksi (469 MPa) will be considered. A description of the reinforced concrete slab and associated modeling is presented next. Interested readers should review the associated web site for additional details. The overall concrete slab dimensions are 64×33.75×4 inches (1625.6×958.85×101.6 mm) with a single layer of reinforcement, as shown in Figure 1, on the side of the slab away from the blast. The concrete slab fixture consists of a steel frame with front and back steel cross supports at the long ends of the slab. Figure 2 shows the final assembly of the blast side fixture over the concrete slab. The slab is mounted in the depicted removable end of a large air blast simulator.
Kasper Cramon Jørgensen, Vivian Swan (NIRAS A/S)
This study investigates the perforation process of armour-piercing projectiles on commercially available high-strength aluminium. A LS-DYNA® model is developed with thick target plates of aluminium alloy 7075-T651 and an incoming 7.62 mm armour-piercing projectile with an impact velocity of 850 m/s. A numerical formulation combining classic Lagrangian finite elements with an adaptive mesh algorithm is utilized to overcome large deformation challenges and more accurately predict failure mechanisms. Both aluminium target and projectile have been modelled as deformable with a modified version of the Johnson-Cook strain-rate and temperature dependent plasticity model, based on input parameters from literature. Main model results include projectile residual velocity after target perforation and prediction of target failure mechanism. The model results are validated against experimental results from live ballistic tests and a sensitivity study is carried out to identify influential material model parameters.
Zeng-Chan Zhang, Grant O. Cook, Jr. & Kyoung-Su Im (LSTC)
CESE compressible fluid solver is one of the new solvers in LS-DYNA R7.0. This solver is based on the space-time conservation element and solution element (CE/SE) method, originally proposed by Chang . The CE/SE method has many non-traditional features, such as (i) both local and global flux conservation are well maintained in space and time; (ii) shock waves can be captured automatically without using Riemann solvers or special limiters, etc. For more details about the CE/SE method, see the references [1, 2, 3]. This method is suitable for high-speed flows, especially with complex shock waves. In the past, the CESE method has been widely used in many different CFD-related areas, e.g., shock/acoustic wave interaction, detonation waves, cavitation, chemically reacting flows, etc.
Roger Grimes, Cleve Ashcraft (LSTC)
The most egregious serial bottleneck for Large Implicit Mechanics modeling for distributed memory parallel execution, independent of the application package, is the sparse matrix ordering for the direct matrix solution. LSTC is developing a new distributed memory ordering algorithm that is at least as effective as the serial algorithm METIS but is a fully scalable implementation. We will give an overview of the algorithm and the impact on some benchmark problems.
Kelly S. Carney, Samuel A. Howard (NASA Glenn Research Center), Brad A. Miller (Harding University, Searcy), David J. Benson (University of California San Diego, La Jolla)
Non-linear, dynamic, finite element analysis is used in various engineering disciplines to evaluate high-speed, dynamic impact and vibration events. Some of these applications require rotation of some elements relative to other elements with various levels of constraints. For example, bird impacts on rotating aircraft engine fan blades is a common analysis done using this type of analysis tool. Traditionally, rotating machines utilize some type of bearing to allow rotation in one degree of freedom while offering constraints in the other degrees of freedom. Many times, bearings are modeled simply as linear springs with rotation. This is a simplification that is not necessarily accurate under the conditions of high-velocity, high-energy, dynamic events such as impact problems. For this reason, it is desirable to utilize a more realistic non-linear force-deflection relationship characteristic of real bearings to model the interaction between rotating and non-rotating components during dynamic events. The present work describes a rolling element bearing model developed for use in non-linear, dynamic finite element analysis. This rolling element bearing model has been implemented in LS-DYNA as a constraint, *CONSTRAINED_BEARING.
Chirag S. Shah, Suraush Khambati, Brock Watson, Nishant Balwan, Zaifei Zhou, Fuchun Zhu, Shiva Shetty (Humanetics Innovative Solutions, Inc.)
Finite Element (FE) models of Anthropomorphic Test Device (ATD), commonly known as crash test dummies, have become increasingly applicable in automotive safety. A variety of ATDs models are widely used in many areas such as restraint development, automotive crashworthiness, occupant safety and other automotive environment related applications. With the increase in cost effectiveness of computational power, progressively complex and detailed computer models of ATDs have become more realistic in recent years. There has been growing demand for these models due to the inherent benefits of reduced cost and time in the product development cycle. The presented paper highlights the development process of two of such highly detailed frontal impact ATD models namely: THOR-M 50th and Harmonized Hybrid III (HIII) 50th in the LS-DYNA FE code. Both these dummy models represent anthropometry of a 50th percentile adult male. The current work describes the model development process and a controlled loading case for each of the dummies to illustrate the predictive capabilities of both models. The geometries and inertial properties for both dummy models are obtained from available drawings and hardware. The model connectivity and structural integrity are inspected by experiments and verified against hardware. Material tests have been conducted for all critical materials, enabling characterization using the latest material modeling techniques. The model’s material properties are implemented from physical test data after numerical parameter extraction and verification through coupon simulations, using available material cards. All the injury output sensors and instrumentation in these models are developed and implemented based on all possible instrumentation information in hardware. These models are then validated against a variety of component, sub-assembly, and full dummy level load cases, as a key for developing reliable models that meet industry expectations. A detailed validation case of the thorax is presented for the Harmonized HIII 50th and a neck validation case is presented for the THOR-M 50th dummy. The current development status has shown very reasonable predictive capabilities of these two models as evident in the illustrated loading conditions which range from component to full dummy level.
Veronika Effinger, André Haufe (Dynamore GmbH), Paul DuBois (Consultant), Markus Feucht (Daimler AG), Manfred Bischoff (University of Stuttgart)
Lightweight design is one of the major principles in automotive engineering and has made polymer materials to inherent parts of modern cars. In addition to their lightweight potential thermoplastics, elastomers, fabric and composites also incur important functions in passive safety. In the age of virtual prototyping, assuring these functions requires the accurate modeling of the mechanical behavior of each component. Due to their molecular structure, polymer materials often show viscoelastic characteristics such as creep, relaxation and recovery. However, considering the general state of the art in crash simulation, the viscoelastic characteristics are mainly neglected or replaced by viscoplastic or hyperelastic and strain rate dependent material models. This is either due to the available material models that are often restricted to linear viscoelasticity and thus cannot model the experimental data or due to the time consuming parameter identification. In this study, a nonlinear viscoelastic material model for foams is developed and implemented as a user material subroutine in LS-DYNA®. The material answer consists of an equilibrium and a non-equilibrium part. The first one is modeled with a hyperelastic formulation based on the work of Chang  and formerly implemented as *MAT_FU_CHANG_FOAM in LS-DYNA (*MAT_083). The second one includes the nonlinear viscoelastic behavior following the multiple integral theory by Green and Rivlin . The polyurethane foam Confor® CF-45 used as part of the legform impactor in pedestrian safety was chosen for its highly nonlinear viscoelastic properties to test the presented approach. The investigation shows the ability of the method to reliably simulate some non-linear viscoelastic phenomena such as saturation.
M. Souli, R. Messahel (University of Lille), B. Cohen (EDF UTO), N. Aquelet (LSTC)
In the nuclear and petroleum industry, supply pipes are often exposed to high pressure loading which can cause to the structure high strains, plasticity and even in the worst scenario failure. Fast hydraulic transient phenomena such as Water Hammers (WHs) are of this type. It generates a pressure wave that propagates in the pipe causing high stress. Such phenomena are of the order of few msecs and numerical simulation can offer a better understanding and an accurate evaluation of the dynamic complex phenomenon including fluid-structure interaction, multi-phase flow, cavitation effects … For the last decades, the modeling of phase change taking into account the cavitation effects has been at the centre of many industrial applications (chemical engineering, mechanical engineering, … ) and has a direct impact on the industry as it might cause damages to the installation (pumps, propellers, control valves, …). In this paper, numerical simulation using FSI algorithm and the two One-Fluid Cavitation models “Cut-Off” and “HEM” of WHs including cavitation effects is presented.
Yuli Huang, Jack Yiu, Jack Pappin, and Richard Sturt (Arup) Julian S. H. Kwan and Ken K. S. Ho (Hong Kong SAR Government)
H. Kim, J. Gould (Edison Welding Institute), J. Shang (American Trim), A. Yadav, R. Meyer (Caterpillar Inc), Pierre L’Eplattenier (LSTC)
In this study, numerical simulations on electro-magnetic welding (EMW) were conducted for dissimilar materials joint of the ring-shaft assembly. LS-DYNA® electromagnetism module was adopted to simulate the EMW process. Simulation results were correlated with the EMW experimental works with two different joint designs, single and double flared lap joint. Two different materials, aluminum 6061-T4 and copper, C40, were used for the driver ring material on the stationary steel shaft. LS-DYNA simulation model was used to investigate the effects of impact angle and velocity on surface-layer bonding and joining efficiency of the driver ring on a steel shaft. Analytical modeling was also conducted to estimate the magnetic pressure between the coil and the ring. Experimentally, a 90-KJ machine was used at different energy levels. From these experiments, the double flared lap joint showed better joint efficiency and the copper showed better adhesion than aluminum at same energy levels. The performance of joint was evaluated by push-off testing. A double flared copper ring at 81-KJ gave the best performance of joint, and exceeded the required axial thrust load requirement. From the metallographic analysis, the interface of joint did not show the metallurgical bonding, however, strong mechanical interlocking was achieved. This study demonstrates the viability of EMW process for dissimilar material joining.
Ronald F. Kulak (RFK Engineering Mechanics Consultants LLC)
Rollover crashes are responsible for many occupant injuries and fatalities. Rollover crash fatalities account for 36 percent of total fatalities for passenger cars and light trucks. Front seat occupants are vulnerable to head, neck and thoracic injuries resulting from impact with the collapsing roof structure. Modeling and simulation on parallel computing platforms using state-of-the-art software – such as LS-DYNA® – is an attractive and economical approach for studying the structural responses of the vehicle and occupant to rollover events. This paper presents simulations of rollover events of a full-sized sedan subjected to several initial vehicle orientations and front occupant positions. The National Crash Analysis Center database provided the finite element model for the full-size sedan. The front-seat occupant model is the Hybrid III finite element model developed by Livermore Software Technology Corporation, which represents the 50% male anthropomorphic test device (ATD). Thus, this study makes use of a single software platform for analyses of both the vehicle and occupant – leading to efficient computations. The current work focused on Single Event Single Rollovers (SESR). Several case studies are presented, and one case simulated a previously performed test using the Controlled Rollover Impact System (CRIS). The first case (far side impact) matched the CRIS Test 51502 initial release conditions, and the numerical simulations match the kinetic conditions when the vehicle contacted the ground – as calculated by rigid body dynamics. The second case looked at near side impact, and the third case looked at far side impact but with a 10 degree pitch angle. Results show that the largest neck forces occur for near side impact. Comparison of the first case simulation results with CRIS Test 51502 is examined for suitability of validating the finite element models to rollovers.
Nils Karajan (DYNAmore GmbH), Zhidong Han, Hailong Teng, Jason Wang (LSTC)
The goal of this contribution is to discuss the assumptions made when modeling granular media with the discrete-element method (DEM). Herein, particular focus is drawn on the physical interpretation of the involved material parameters of the DEM in LS-DYNA. Following this, the influence of each parameter on the bulk behavior of granular media is investigated and different possibilities to estimate these parameters are presented.
Filipe Andrade, Andre Haufe (DYNAmore GmbH), Markus Feucht (Daimler AG)
As a consequence of the worldwide tendency in reducing CO2 emissions by producing lighter and more energy-efficient products, the demand for accurate predictions regarding material behavior and material failure has greatly increased in recent years. In particular in the automotive industry, there is also an increasing interest in effectively closing the gap between forming and crash, since the forming operations may highly affect the crashworthiness of the produced parts. In this scenario, a correct depiction of material mechanical degradation and material fracture seems indispensable. Currently, there are several models implemented in LS-DYNA which have been developed to deal with material damage and failure. Many of them are complete constitutive models which consider elasto-plasticity coupled with damage formulations as well as with embedded failure criteria (e.g., *MAT_015, *MAT_052, *MAT_081, *MAT_104, *MAT_120, *MAT_153, among others). Alternatively, LS-DYNA also makes possible the definition of failure and damage through the keyword *MAT_ADD_EROSION, where the user can choose different failure models and fracture criteria which are, in turn, coupled with the selected plasticity model in an ad-hoc fashion. In this context, GISSMO (Generalized Incremental Stress-State dependent Damage Model) and DIEM (Damage Initiation and Evolution Model) are good candidates for the task of predicting ductile failure using LS-DYNA. However, many users still seem to have difficulties in using these models, meanwhile other users, who already master either GISSMO or DIEM, feel somewhat insecure in employing the concurrent model. These difficulties arise mainly because GISSMO and DIEM have been conceived following quite different interpretations of the phenomena that influence failure. For instance, in GISSMO the user has to input a failure curve as a function of the triaxiality (and also of the Lode parameter, in the case of solid elements) where this curve is used for the nonlinear accumulation of damage. This strategy intrinsically takes the strain path change into account, for which a numerical calibration based on experimental data is required. Furthermore, an instability curve may also be defined in GISSMO, where in this case, if instability achieves a critical value, the stresses are assumed to be coupled with damage, leading to a ductile dissipation of energy upon fracture. DIEM, on the other hand, allows the user to define multiple damage initiation indicators which evolve simultaneously. For example, the user can define a normal and a shear failure initiation criterion, the former as a function of triaxiality, the latter depending on the so called shear stress function. Additionally, a forming limit curve (FLC) can also be input in DIEM, where this criterion also evolves along the other two failure initiation criteria. The different damage initiation criteria can then be combined in a global damage evolution rule. Similarly to GISSMO, a certain number of experiments is required in order to properly fit the parameters necessary for DIEM. This contribution is an attempt to compare and better understand the differences between GISSMO and DIEM. In this respect, the main differences between both models and how they are intended to predict failure will be comprehensively discussed. Additionally, the calibration of a dual-phase steel using GISSMO and DIEM will be used to better highlight the differences between the models and how these are reflected in the final parameter fitting.
Xiaomin Zeng, Xiongqi Peng (Shanghai Jiao Tong University), Hongsheng Lu (Shanghai Hengstar Technology Co. Ltd), Edmondo Di Pasquale (SimTech Simulation et Technologie)
Pedestrian head impact with bonnet is one of the major causes for pedestrian severe injury or fatality. This paper proposes a multidisciplinary design optimization method for bonnet inner based on pedestrian head protection along with stiffness requirements. The static stiffness and headform collision procedure with regard to a particular industrial bonnet are analyzed. Parametric design and optimization analysis of this bonnet are carried out. Optimization solution significantly achieves better head protection effect under the premise of meeting the stiffness requirements, which validates the feasibility of this multidisciplinary optimization method and provides an approach for the optimal design of engine bonnet inner. This work shows the importance of a simultaneous approach of different disciplines in bonnet design.
Hailong Teng, Jason Wang (LSTC)
This paper presents a particle blast method (PBM) to describe blast loading. The PBM is an extension of corpuscular method (CPM), which is coarse-grained multi-scale method developed for ideal gas dynamics simulation. It is based on the kinetic molecular theory, where molecules are viewed as rigid particles obeying Newton’s laws of mechanics, while each particle in the particle method represents a group of gas molecules. Pressure loading on structures is represented by particle-structure elastic collisions. The corpuscular method has been applied to airbag deployment simulation where the gas flow is slow. For blast simulation where gas flow is extremely high, the particle method has been improved to account for the thermally non-equilibrium behavior. Furthermore, to better represent gas behavior at high temperature, co-volume effects have been considered. The particle blast method could be coupled with discrete element method, make it possible to model the interaction among high explosive detonation products, the surrounding air, sand and structure.
Hadar Raz (Plasan Ltd.)
For crash and blast tests of vehicles and sub-assemblies, simulations play an important role in the prediction of the test results. Some of the most important results are the occupants’ injury criteria, which are calculated by simulating ATDs and their various joints, accelerometers, etc’. Often in a simulation/test there are few ATDs, and there is an increasing demand for post-processing of the injury criteria in an automated way, as well as correlating the results between simulation and test, thus enabling easier calibration of the simulation. We present PC3 (Plasan Criteria Computation and Comparison), a software tool developed by Plasan, which enables easy calculation of simulation and test ATD results, and correlation of said results. Currently the program is able to read simulation data from LS-DYNA® binout database, and various test databases, such as ISO text files, CSV files, HDF5 database files and some others. An example of criteria calculation for blast simulation and test data will be shown, along with correlation between the two.
Peter H. Foss (General Motors Global Research & Development)
As part of a cooperative development project between General Motors, BASF and Montaplast, a glass reinforced nylon oil pan was designed, analyzed, molded and tested. The oil pan was molded from BASF’s Ultramid® B3ZG7 OSI, an "Optimized for Stone Impact" grade of impact modified 35% short glass filled PA6. One of the development tests run on the pans was a drop impact test. In this report we will compare the predicted and experimental impact response using Digimat and LS-DYNA® with an anisotropic elastic-viscoplastic material model with failure. The Digimat material model was reverse engineered from high-rate tensile stress-strain data provided by BASF.
Eyal Rubin, Yoav Lev (RAFAEL Advanced Defense Systems LTD.)
A steel band is tightened around a thin walled steel cylinder. The assembly is exposed to different dynamic loadings including shock and vibration. While tightening, the circumferential stresses developed in the band, decrease as a function of the distance from the bolts and the value of the coefficient of friction between the band and the cylinder. The cylinder elasticity also affects the amount of force distribution in the band. A rigid cylinder will result in a maximal distribution of internal tension forces in the band. Experiments show that dynamic loadings, such as shock and vibrations, release the initial preload of the tightening bolts, and average the distribution of internal tension forces in the band. The extent of the change in the internal forces distribution depends on the level of the dynamic loading. While the motivation of the work was to find a lower boundary to the tightening force, a severe shock was chosen to demonstrate this Phenomenon. As a result from the severe shock, the internal tension forces at different cross sections converged to the same final uniform force. The level of this final force varies, depending on the coefficient of friction. The maximum possible release of the internal tension forces in the band, as a function of the coefficient of friction between the cylinder and band, and the rigidity of the cylinder, was determined using LS-DYNA® explicit simulation. This method can be used to determine the initial tightening force of any assembly, in order to assure that it stands dynamic environmental conditions.
Bazle Z. (Gama) Haque, John W. Gillespie Jr. (University of Delaware)
Performing experiments in numerical space and predicting accurate results are the main research focus of many computational mechanicians. These goals may in general sound challenging, however, makes perfect sense in cases where experiments are not possible, e.g., landing on Mars, sea waves impacting marine structures, crash landing of space shuttle, etc. Composite damage modeling plays a vital role in designing composite structures for damage tolerance, energy dissipating crash, impact, ballistic, and blast applications. A progressive composite damage model MAT162 is developed by Materials Sciences Corporation and further modified by the authors and implemented in explicit finite element analysis code LS-DYNA. A total of thirty-four material properties and parameters are required to define such a material model. Besides the ASTM standard test methods for determining the elastic and strength properties, the authors have developed a low velocity impact methodology in determining the rate insensitive model parameters. Recently, model validations with depth of penetration and ballistic experiments have been performed to determine the rate sensitive model parameters. These validated model parameters are used to predict composite damage and resistance behavior of composite structures made from plain-weave plain weave S-2 glass/SC15 composites under quasi-static, low velocity impact and crush, ballistic, and blast loading conditions. Analysis procedure and results of these numerical experiments will be presented.
Cezary Bojanowski (Argonne National Laboratory), Marcin Balcerzak (Warsaw University of Technology)
The computational analysis of engineering structures under blast loads faces three fundamental problems: (i) reliable prediction of blast loads imposed on structures, (ii) correct representation of material behavior, and (iii) global analysis of large scale structures. Despite the recent developments in Finite Element (FE) codes like LS-DYNA® and advancements in computational power, addressing all of these issues in a single simulation is not a straightforward task. In this paper, LS-DYNA capabilities were utilized to simulate the transient global response of a long span cable stayed bridge subjected to blast loading over the deck and to evaluate localized damage to the deck structure. Described in detail is the development of a global FE model of the Bill Emerson Memorial Bridge – a cable-stayed bridge crossing over the Mississippi River near Cape Girardeau, Missouri. The global model takes into account the structural details of the deck, support columns and the pretension in the stay cables. It was partially validated by comparing the calculated natural frequencies with those previously extracted by the Missouri Department of Transportation from data recorded during the 2005 earthquake of M4.1 on a Richter scale (Assessment of the Bill Emerson Memorial Bridge, Report No. OR08-003, September 2007). A detailed model of the central section of the deck was developed to simulate localized damage. Boundary conditions on the detailed model were applied through a sub-modeling technique based on the analysis of the global simplified model. The results show that a detonation of explosives of a typical size of passenger car and van bomb on a traffic lane in the mid span of the deck is not likely to cause a collapse of the bridge. The vibrations in the stay cables do not lead to yielding of the steel in the strands. The simulation of the local damage shows that – for the chosen van location – the blast may perforate the deck and deform the cross beam. The extent of the damage, however, depends greatly on the assumed erosion criteria.
Bill Loewe (Panasas, Inc.)
As HPC continues its growth with Linux clusters using multi-core processor architectures, the I/O requirements further increase with higher-fidelity CAE modeling and workflow demands. This paper examines the parallel scalability characteristics of LSTC’s Finite Element Analysis software LS-DYNA for up to 288 processing cores for implicit mechanics simulations that have high I/O demands. The motivation was to quantify the performance and scalability benefits of parallel I/O in FEA software using a parallel file system, compared with both local storage and conventional NFS for implicit mechanics cases. This study was conducted on a Linux Intel Xeon cluster with a Panasas PanFS parallel file system and using a benchmark input provided by Roger Grimes from LSTC. For this study, relevant models used were based on current customer practice to demonstrate that LS-DYNA with parallel I/O can show a significant performance advantage and corresponding reduction in job overall time for advanced implicit simulations.
Karen E. Jackson, Justin D. Littell (NASA Langley Research Center), Edwin L. Fasanella (National Institute of Aerospace)
In 2010, NASA Langley Research Center obtained residual hardware from the US Army’s Survivable Affordable Repairable Airframe Program (SARAP), which consisted of a composite fuselage section that was representative of the center section of a Black Hawk helicopter. The section was fabricated by Sikorsky Aircraft Corporation and was subjected to a vertical drop test in 2008 to evaluate a tilting roof concept to limit the intrusion of overhead mass items, such as the rotor transmission, into the fuselage cabin. As a result of the 2008 test, damage to the hardware was limited primarily to the roof. Consequently, when the post-test article was obtained in 2010, the roof area was removed and the remaining structure was cut into six different types of test specimens including: (1) tension and compression coupons for material property characterization, (2) I-beam sections, (3) T-sections, (4) cruciform sections, (5) a large subfloor section, and (6) a forward framed fuselage section. In 2011, NASA and Sikorsky entered into a cooperative research agreement to study the impact responses of composite airframe structures and to evaluate the capabilities of the explicit transient dynamic finite element code, LS-DYNA®, to simulate these responses including damage initiation and progressive failure. Finite element models of the composite specimens were developed and impact simulations were performed. The properties of the composite material were represented using both a progressive in-plane damage model (Mat 54) and a continuum damage mechanics model (Mat 58) in LS-DYNA. This paper provides test-analysis comparisons of time history responses and the location and type of damage for representative I-beam, T-section, and cruciform section components.
Edwin L. Fasanella (National Institute of Aerospace), Karen E. Jackson, Justin D. Littell (NASA Langley Research Center), Michael D. Seal (Analytical Mechanics Associates, Inc.)
NASA Langley Research Center obtained a composite helicopter cabin structure in 2010 from the US Army’s Survivable Affordable Repairable Airframe Program (SARAP) that was fabricated by Sikorsky Aircraft Corporation. The cabin had been subjected to a vertical drop test in 2008 to evaluate a tilting roof concept to limit the intrusion of overhead masses into the fuselage cabin. Damage to the cabin test article was limited primarily to the roof. Consequently, the roof area was removed and the remaining structure was cut into test specimens including a large subfloor section and a forward framed fuselage section. In 2011, NASA and Sikorsky entered into a cooperative research agreement to study the impact responses of composite airframe structures and to evaluate the capabilities of the explicit transient dynamic finite element code, LS-DYNA®, to simulate these responses including damage initiation and progressive failure. Most of the test articles were manufactured of graphite unidirectional tape composite with a thermoplastic resin system. However, the framed fuselage section was constructed primarily of a plain weave graphite fabric material with a thermoset resin system. Test data were collected from accelerometers and full-field photogrammetry. The focus of this paper will be to document impact testing and simulation results for the longitudinal impact of the subfloor section and the vertical drop test of the forward framed fuselage section.
Hyunwook Kim (Memorial University of Newfoundland)
A laboratory scale compressive cone-shaped ice experiments were performed, and a numerical simulation model using LS-DYNA was developed. Modified material properties were applied based on a crushable foam model (MAT 63) as the ice properties. To simulate a saw-tooth pattern which is commonly observed through experiments in ice, an additional function of failure criteria, which is maximum principal stress, was included. Results of the experimental and numerical simulation were compared and represented a good agreement. The proposed numerical simulation model was extended to a larger scale and verified.
V. Mamutov (St. Petersburg State Polytechnical University), S. Golovashchenko (Ford Motor Company), A. Mamutov (Oakland University)
The method of simulating of expansion of plasma channel during high-voltage discharge in water at electro-hydraulic forming (EHF) is developed using finite-element software package LS-DYNA® 971. The energy input into the plasma channel from the discharge circuit is taken into account. The energy deposit law is calculated from the experimentally obtained pulse current and voltage in the discharge circuit measured between the electrodes. The model of discharge channel is based on assumption of adiabatic channel expansion when the energy introduced into the channel from the external electrical circuit is only spent on increasing the internal energy of plasma and on the work of plasma expansion. Water is simulated as an ideally compressible liquid with bulk modulus of K = 2.35 GPa. The cavitation threshold for the liquid is defined as 0.1 MPa. The interaction between the channel and the water is simulated using Arbitrary Lagrange Eulerian (ALE) technique. The deformable blank is simulated using shell elements. The results of simulation of two variants of the discharge chamber are presented: for the long cylindrical chamber with long axisymmetrical discharge channel and for the compact chamber of arbitrary shape with short discharge distance. The developed numerical method is verified by comparing the results of simulation with the results from the test simulation, which is a one-dimensional axisymmetric finite-difference based problem with the same parameters. It is also verified by comparing the results of simulation and the experimentally measured pulse pressure in the discharge chamber with a known function of energy input.
Mikael Schill (DYNAmore Nordic AB), Eva-Lis Odenberger (Industrial Development Centre in Olofström AB)
Predicting the finished geometry of a part is a major issue for the manufacturing industry. This is a complex task, especially if the manufacturing involves several types of processes. In order to succeed, the complete manufacturing process has to be included in the simulation. For sheet metal forming, this has been done for quite some time where trimming, forming and springback are simulated in consecutive order. However, there are other manufacturing processes which affect the geometry of the finished part. In this paper, a welding process is added to the manufacturing process chain. The welding simulation is done using the novel material model *MAT_CWM with ghost element functionality. The aim of the paper is to investigate how the different process stages affect the final geometry of the part and how this is efficiently and accurately simulated with LS-DYNA. Further, an attempt is made to improve the part tolerance by springback compensation of the forming tools accounting for deviance from both springback and weld deformation.
Ken-An Lou (ArmorWorks), David Bosen, Kiran Irde, Zachary Blackburn (ShockRide)
Hybrid III Anthropomorphic Test Dummy (ATD) is primarily validated for frontal impacts from physical sled tests in an automotive incident but not for a military vehicle incident related to mine blast vertical impacts. Vertical drop tests were conducted using Hybrid III 50th percentile ATD. The purpose of conducting these tests was to identify which LSTC dummy model shows the best correlation with the test results. This paper presents the modeling correlation between LSTC’s 50th percentile RIGID, FAST, and DETAILED dummy models. A rigid seat without seat cushion was used in the drop tests so the surroundings the dummy interacted with during the test were very predictive. A total of three drop tests from the same drop height were completed to ensure consistency and repeatability of the test data. The simulation was correlated to the test data for occupant responses.
Shinya Hayashi (JSOL Corporation), Richard Taylor (Ove Arup & Partners International Limited)
Computer simulation is playing an increasingly important role in the design, development and application of airbag safety systems. As folding patterns and airbag structures become more and more complex, users are turning to simulation based folding solutions to generate accurately folded models in a short space of time. To meet this demand, a new software tool called JFOLD has been developed by JSOL Corporation to enable successful airbag folding using LS-DYNA®. JFOLD’s intuitive and interactive system guides the user through the folding steps using flow-chart graphics, interactive tool positioning/resizing, tool motion control, animation preview and more. This paper introduces the new capabilities of JFOLD Version 2 and demonstrates various folding examples. JFOLD runs inside the powerful and popular pre-processor Primer.
Eric Strong, Hubert Lobo (Matereality), Brian Croop (DatapointLabs)
LS-DYNA contains a wealth of material models that allow for the simulation of transient phenomena. CAE Modeler is a generalized pre-processor software used to convert material property data into material parameters for different material models used in CAE. In a continuation of previously presented work, we discuss the extension of the CAE Modeler software to commonly used material models beyond MAT_024. Software enhancements include advanced point picking to perform extrapolations beyond the tested data, as well as the ability to fine-tune the material models while scrutinizing the trends shown in the underlying raw data. Advanced modeling features include the ability to tune the rate dependency, as well as the initial response. Additional material models that are quite complex and difficult to calibrate are supported, including those for hyperelastic and viscoelastic behavior. As before, the written material cards are directly readable into the LS-DYNA software, but now these can also be stored and cataloged in a material card library for later reuse.
Frank Marrs, Mike Heiges (Georgia Tech Research Institute)
This paper presents the results of an effort to correlate an LS-DYNA® simulation of a buried mine blast with published experimental test data. The focus of the study was on simulating the effects of soil moisture content on the blast characteristics. A mathematical model for sand is presented that is based on several previously proposed models. The simulation correlates well with the results of a mine blast experiment, thus validating the material model for sand at varying levels of saturation. The model provides an excellent baseline for blast simulations of buried mines and a soil material model that can be expanded to include higher fidelity modeling, different soil types, and real-world applications.)
Zhe Cui, Yun Huang (LSTC)
Volker Steininger, Xinhai Zhu, Q. Yan, Philip Ho (Tiwa Quest AG, LSTC)
In recent years the spring back calculation of a sheet metal part after forming has achieved a high accuracy. Today we are able to calculate the spring back amount after all forming operations of the part including trimming, piercing and reforming. But the spring back of a single sheet metal part is only the first step to solve the problem. What matters at the end is the spring back of the assembly or the sub-assembly of multiple sheet metal parts. This paper describes a GUI for an efficient setup of a spring back calculation of multiple sheet metal parts, taken into account the complete forming history of the parts. It will be shown how to position the sheet metal parts, how to define an assembly sequence and joining method and the hemming process of outer and inner panels and how to launch the LS-DYNA® simulation.
Willem Roux (LSTC)
The LS-TaSC product status is presented. The current capabilities are discussed together with illustrative examples and release dates. In addition, the current development directions, such as new capabilities and CAE integration, are also revealed.
Jun-Ku Lee, Hyun-Cheol Kim (Theme Engineering, Inc.)
Recently in the field of sheet metal forming, servo press, which can control the speed and position of the tool using a servo motor, is attractive method. Development process of servo press method is accelerated as the capacity of servo motor become bigger. In the future, it’s expected as a great alternative plan to replace conventional press method in order to improve quality, increase productivity, maintain integrity of tools, and reduce energy consumption. Motion control in servo press method has to be effectively optimized depending on the shape and characteristics of the material. However, in the industrial field, the controls of motion relied on experience or intuition of most skilled worker, so the workers can’t avoid many trials and errors to find the optimized motion. We try to implement the servo press method using a finite element analysis with LS-DYNA® in order to shorten the trials and errors . And furthermore, we try to find optimal motion with LS-OPT® . The front side member model of Numisheet 2011 BM03 was chosen for the analysis. We carefully consider stress relaxation and time scaling in order to implement servo press method. Then, we try to compare the following three cases to look into utility of optimized motion for the formability and productivity. – Conventional press method – Conventional press method controlled by speed – Servo press method Finally, we hope that the application of LS-OPT could be effectively used for the optimization of servo press method.
Changqing Du, Kaiping Li (Chrysler Group LLC), Xiaoming Chen (U.S. Steel Corporation), Yuzhong Xiao, Xinhai Zhu (LSTC)
In this study, the simulation and formability prediction of the DP600 Steel Revers Draw in the NUMISHEET 2014 Benchmark 1 is conducted using LS-DYNA®. The combinations of the material models and element formulations are evaluated for better strain path correlations between the simulation and the measurement at the specified points. Various input factors are considered in this study, including different material model and element type, mesh sizes, integration points and locations. In addition to the conditions given in the benchmark description, extra factors such as the friction effects and springback after drawbead forming process are also considered. The simulation results show that the properly selected yield function is critical for the stain path predictions to be in better agreement with the experimental measurements under such loading condition. In simulation the Formability-Index method is applied to determine the forming limit strains. With this method, the predicted limit strains of the on-set necking points, as well as the locations are compared with the measurement results reported in Benchmark 1 Analysis.
Robert K. Goldberg and Kelly S. Carney (NASA Glenn Research Center), Paul Du Bois (George Mason University), Canio Hoffarth, Joseph Harrington, Subramaniam Rajan (Arizona State University), Gunther Blankenhorn (LSTC)
Several key improvements in the state of the art have been identified by the aerospace community as desirable for inclusion in a next generation material model for composite materials to be incorporated within LS-DYNA®. Some of the specific desired features that have been identified include the incorporation of both plasticity and damage within the material model, the capability of using the material model to analyze the response of both three-dimensional solid elements and two dimensional shell elements, and the ability to simulate the response of composites composed with a variety of composite architectures, including laminates, weaves and braids. In addition, a need has been expressed to have a material model that utilizes tabulated experimentally based input to define the evolution of plasticity and damage as opposed to utilizing discrete input parameters (such as modulus and strength) and analytical functions based on curve fitting. To address these needs, a multi-institution consortium has been formed to develop and implement a new composite material model within LS-DYNA. To date, the model development efforts have focused on creating and implementing an appropriate plasticity based model. Specifically, the Tsai-Wu composite failure model has been generalized and extended to a strain-hardening based orthotropic plasticity model with a non-associative flow rule. The coefficients in the yield function are determined based on tabulated stress-strain curves in the various normal and shear directions, along with selected off-axis curves. The non-associative flow rule is used to compute the evolution of the effective plastic strain. Systematic procedures have been developed to determine the values of the various coefficients in the yield function and the flow rule based on the tabulated input data.
Devon Downes, Manouchehr Nejad Ensan (Aerospace Portfolio, National Research Council), Amal Bouamoul (Defence Research & Development Canada–Valcartier)
Fragments of aluminum impacting on Composition B explosive encased in rolled homogenous armour (RHA) steel were investigated through the LS-DYNA®. The investigation focused on shock to detonation simulations of Composition B, with the objective of determining both the critical velocity which would generate a shockwave strong enough to cause detonation of the explosive, as well as the resulting pressure profile of the detonation wave. Detonation scenarios at low, intermediate and high impact velocities were investigated. It was observed that at low impact velocity the explosive failed to detonate. At intermediate velocities, detonation was due to the development of localized hot spots caused by the compression of the explosive from the initial shockwave. Detonation was also caused by pressure waves reflecting against the casing of the explosive leading to the so-called sympathetic detonation. At high impact velocity, initiation of the explosive was caused by the initial incident pressure wave located immediately behind the top casing/explosive interface.
Ofir Shor and Reza Vaziri (The University of British Columbia)
The increasing use of laminated composite materials in advanced industrial applications requires the ability to predict their behavior under the expected service loads. Laminated composite materials exhibit various damage and failure mechanisms, which can cause a degradation of their mechanical properties and lead to catastrophic structural failure. The debonding of adjacent laminate layers, also known as delamination, is considered to be one of the most dominant damage mechanisms in the failure of composite laminates; hence it is important to have numerical analysis tools that are able to predict its initiation and growth. Although methodologies to simulate delamination in composite materials exist, they are limited to small-scale models and are therefore not suitable for large-scale applications of practical significance. A new method is presented here that allows simulation of this type of failure mode in large-scale composite structures. This is achieved by locally and adaptively splitting the structural elements through their thickness, while introducing cohesive zones in regions where delamination has the potential to initiate. The delamination damage can thus propagate in the structure as the simulation progresses. A mechanical benchmark example (Figure 1) is solved using this approach and the results are verified against those obtained using other numerical methods.
Katharina Witowski, Heiner Müllerschön, Andrea Erhart, Peter Schumacher (DYNAmore GmbH)
This paper deals with topology and topometry optimization of structures under highly nonlinear dynamic loading such as crash using equivalent static loads. The basic idea of the “Equivalent Static Load”- Method (ESL) is to divide the original nonlinear dynamic optimization problem into an iterative “linear optimization ↔ nonlinear analysis” process. The displacement field of the nonlinear dynamic analysis is transformed to equivalent linear static loads for a variety of time steps. This leads to an optimization with multiple linear static load cases. In an outer loop the nonlinear analysis is repeated to correct and adapt the displacement field. There are several previous papers on the idea of equivalent static loads, e.g. Shin MK, Park KJ, Park GJ: Optimization of structures with nonlinear behavior using equivalent load. Comp. Meth. Appl. Math.,196, p.1154-1167, 2007. This paper reports about experiences in the application of the ESL methodology on industrial problems from the automotive industry. For the nonlinear dynamic analysis LS-DYNA® is used, for linear topology and topometry optimization GENESIS from Vanderplaats R&D is applied. The investigations have been performed within a research project, founded by the association BMBF, with several partners from German automotive companies. On the application of the method on large scale problems numerous problems are encountered. Setting up a fully automated and robust process on an HPC cluster with nested linear and nonlinear finite element analysis and optimization for multiple load cases was a challenging task. The general objective of the investigations was to evaluate the suitability of the method for different types of crash and impact problems. The appraisement is with respect to quality and usability of the results and with respect to the numerical costs.
Sebastian Stahlschmidt, Alexander Gromer, Reuben D´Souza, Ulrich Franz (DYNAmore GmbH)
The FAT and PDB dummy models have been developed for more than a decade. The models are used by almost all OEMs and restraint system suppliers to enhance the passive safety performance of their vehicles. Nevertheless, the PDB is still launching new projects to further enhance the predictability of the ES-2, ES-2re, BioRID-II and WorldSID models. This paper presents the current enhancement projects at a glance. With the increasing quality of the dummy and the restraint system models, additional details related to model assembly have a significant influence on the overall accuracy. Thus, a more advanced pre-processing to assemble the finite element input is required. Such pre-processing might involve a sequence of pre-simulations and might take pre-stresses into account. In some cases the computational effort for the pre-processing exceeds the time needed to simulate the final load case. The paper presents ideas and solutions to simplify and to speed-up the pre-processing of the above mentioned dummy models without a loss in accuracy. The solutions utilize standard pre-processing tools and scripts as well as LS-DYNA® implicit and explicit time-stepping schemes.
Ming-Pei Lin, Chih-Min Chang, Cho-Hsuan, Tsai, Chia-Hui Tai, Chun-Te Lee (HAITEC)
Occupant simulation is very useful for vehicle restraint system and passive safety development. Since the requirements of safety assessment are more and more demanding, the occupant simulation models have to be more and more accurate. Dummy model is very crucial in occupant simulation, and is difficult for most vehicle manufacturer to build. To purchase a commercial dummy model is a very reasonable and safe choice. Alternatively, LSTC offers LSTC dummy models, which are free for LS-DYNA® users. It is foreseeable that LSTC dummy models may not be as stable as commercial ones. This article describes the usage of LSTC_NCAC Hybrid III 50th dummy in frontal occupant simulation, and try to give a preliminary guideline of usage in vehicle development.
Corey D. Morris, Lisa M. Dangora and James A. Sherwood (University of Massachusetts)
Thermoforming of fabrics is a composite manufacturing process that has the potential to yield quality parts with production costs and cycle times comparable to the fabrication of stamped metal parts. The thermoforming process, illustrated in Figure 1, begins with alignment of the fabric in a rigid frame. Typically, multiple ply layers are simultaneously stamped into the mold to achieve the desired part thickness and mechanical properties. The individual plies can be oriented and aligned in the frame to give the desired directional performance. The loaded frame is transported along shuttle rails to an oven where it is heated until the polymer matrix is hot enough to flow with reasonably low viscosity. Because the traditional materials used for this manufacturing process are fabrics with commingled tows or pre-impregnated sheets, there is no need for a resin infusion step. The frame is moved from the oven to the molding area and aligned between a punch and die; binder plates are conventionally used to apply force around the circumference of the part. The application of pressure to the binder plate induces in-plane forces in the fabric that can reduce wrinkling as it is drawn into the die by the punch. A velocity is then prescribed to the punch to force the ply stack into the die mold. The tools (punch, die and binder plate) are often heated to slow the rate at which the polymer matrix cools. The finished piece assumes the geometry of the die and punch and hardens into a solid part after the matrix has cooled.
F Lancelot (ARUP), YH Zhu2, BH Li, KF Wang, CL Li (Changchun Railway Vehicle Co., Ltd)
The hydraulic gas damper coupler (HGDC) is the most important energy absorbing component in rolling stock crash impact. The HGDC can effectively dissipate crash energy and reduce excessive structural loads at all impact speeds. In this paper, the LS-DYNA material *MAT_HYDRAULIC_GAS_DAMPER_DISCRETE_BEAM (*MAT_070) is used to simulate the HGDC coupler. Using this formulation, both static and dynamic characteristics have been replicated. The validated HGDC models have been incorporated into rolling stock frontal impact simulations. A simple mass-beam representation for the carriages and a full scale detailed model – containing 16 carriages and more than 28 million elements – have successively been analysed. This paper also presents the innovative pre-postprocessing and data storage methods developed by CNR and Arup to handle the very large FE models and results files.
David Karlsson (DYNAmore Nordic AB)
A Mobile Explosive Containment Vessel (MECV) is a chamber for protection against effects caused by explosions and is used to safely secure, contain, transport, store or test explosive materials. The MECV has been tested for a charge equivalent to 8 kg of TNT and strain levels at several positions were measured. These test data were used for comparison and validation of two simulation techniques and if necessary improve the simulation methodology. The first technique uses a separate 2D-axisymmetric MMALE simulation for the explosive blast load calculation and it showed good agreement to the test. In this case, an axisymmetric blast simulation is first made and the pressure is recorded at the fixed boundary. Then an in-house developed program is used to map the blast load to the 3D structure simulation. The second, much more compute intensive technique, is to do a full 3D coupled MMALE simulation of the blast and structure. The second technique lead initially to lower strain levels compared to the test and a more detailed parameter study had to be performed to improve the simulation results. As conclusion, we now have two validated simulation techniques and procedures to make realistic explosive simulations of containment vessels.
Youcai Wu, John E. Crawford, Shengrui Lan, Joseph M. Magallanes (Karagozian & Case)
Many concrete constitutive models are available for use in LS-DYNA®. A thorough validation related to their applicability for the types of problems at hand should be made before application of any of these models. The process for validating a constitutive model includes examining the results produced with the model related to the behaviors it exhibits, gathering a suite of measured data collection pertinent to the problem to be addressed, and comparisons of measured and computed data. This paper addresses issues related to blast response analyses, which include simplification of boundary conditions (such as support condition and contact interfaces), numerical discretization, and material modeling. It was found important that the strain rate effects should be imposed properly since blast loadings usually excite high frequency and high strain rate responses. The impact of boundary conditions was also identified through the numerical studies.
Canio Hoffarth, Joseph Harrington, Subramaniam D. Rajan (Arizona State University), Robert K. Goldberg, Kelly S. Carney (NASA-GRC, Cleveland), Paul Du Bois (George Mason University), Gunther Blankenhorn (LSTC)
A general purpose orthotropic elasto-plastic computational constitutive material model has been developed to accurately predict the response of composites subjected to high velocity impact. The three-dimensional orthotropic elasto-plastic composite material model is being implemented initially for solid elements in LS-DYNA® as MAT213. In order to accurately represent the response of a composite, experimental stress-strain curves are utilized as input, allowing for a more general material model that can be used on a variety of composite applications. The theoretical details are discussed in a companion paper. This paper documents the implementation, verification and validation of the material model using the T800-F3900 fiber/resin composite material.
Swee Hong TAN, Roger CHAN, Jiing Koon POON and David CHNG (Ministry of Home Affairs, Singapore)
Although there are many concrete material models in LS-DYNA®, very few appear to be valid for Multi-Material Arbitrary Lagrangian-Euler (MM-ALE) simulations. From a rudimentary point of view, it makes sense at the first instance, that the typical method of verification via simulating cylinder tests in triaxial and/or uniaxial stress states for Lagrangian format should work for MM-ALE as well. This paper shares the experiences gathered from attempts at simulating cylinder tests involving MM-ALE concrete material models. Useful insights were gained and they would form part of the considerations for future work.