Nicolas Aquelet – Livermore Software Technology Corporation, Mhamed Souli – Laboratoire de Mecanique de Lille
A 2D MMALE (Multi-Material Arbitrary Lagrange Euler) code was implemented in LS-DYNA®. Like the 3D MMALE already available each 2D computational cycle is divided in two steps. First a multi-material version of the two-dimensional shell formulations solves the physical problem on quadrangle meshes during the Lagrangian step. The 2D shell formulations are plane strain and area-weighted axisymmetric. An advection step adapted to the 2D shell approaches follows to control the mesh motion. 2D ALE data of the last cycle can be mapped on 3D ALE mesh. Data are stored in a file defined on the command line after the prompt “map=”. This file is read for the 3D model with the same command line.
Ramesh Edara, Shan Shih – ArvinMeritor Inc., Nasser Tamini, Tim Palmer, Arthur Tang – Engineering Technology Associates
Virtual proving ground (VPG) simulations have been popular with passenger vehicles. VPG uses LS-DYNA® based non-linear contact Finite Element analysis (FEA) to estimate fully analytical road loads and to predict structural components durability with PG road surfaces and tire represented as Finite elements. Heavy vehicle industry has not used these tools extensively in the past due to the complexity of heavy vehicle systems and especially due to the higher number of tires in the vehicle compared to the passenger car. The higher number tires in the heavy vehicle requires more computational analysis duration compared to the passenger car. However due to the recent advancements in computer hardware, virtual proving ground simulations can be used for heavy vehicles. In this study we have used virtual proving ground based simulation studies to predict the durability performance of a trailer suspension frame, tractor suspension frame and combination of both frames with 18 wheel on a given PG event. The virtual proving ground was also used to predict the stress, strain time histories, spindle loads and the component fatigue life for the given PG event.
Yuli Huang, Stephen A. Mahin – University of California at Berkeley
In analysis and design of structures subjected to earthquakes, the cyclic and dynamic nature of the response leads to complications. Material models need to account for cyclic plasticity, including deterioration and eventual failure due to low-cycle fatigue. A cyclic damage plasticity model MAT_DAMAGE_3 (MAT_153, LSTC 2007) is implemented to combine Armstrong-Frederick/Chaboche nonlinear kinematic hardening, isotropic hardening, and Lemaitre isotropic damage evolution based on continuum damage mechanics. By appropriately choosing parameters, this model can reproduce an approximation to the widely-accepted Manson-Coffin low-cycle fatigue rule without of cycle counting. This makes it possible to model the decrease in the material’s ability to deform inelastically. The material model is applied to assess the behavior of a steel structure subjected to deterioration and failure.
Costin Untaroiu, Dan Genovese, Johan Ivarsson, Jeff Crandall – University of Virginia
Bone fractures occur frequently at mid-shaft femoral site during frontal and offset automotive crashes. Because these injuries are expensive, it is crucial to understand the injury mechanisms if this injury is to be prevented. The experimental investigation of femoral shaft tolerance under loading corresponding to real world accidents requires a challenging test setup that allows applying external loads in controlled conditions, mimics the boundary conditions of the femur, and measures the loads at the mid-shaft cross-section of the femur. In addition, the variability of mechanical and structural properties of the specimens complicates the determination of the injury tolerance of the femur under different loading conditions. A numerical alternative is presented in the current study. First, a subject specific finite element model of a femur is developed based on medical images. Then, the parameters of two material models frequently used to approximate the cortical bone properties are identified using the Successive Response Surface Methodology in the ranges reported in the literature. The objective function is defined based on the impact force data recorded during a three-point bending test and its corresponding numerical simulations. The polynomial meta-models implemented in LS-Opt converge at close values of the material parameters suggesting good performance of the heuristic design search in the current identification problem. The femoral tolerance at mid-shaft location is determined using a virtual test setup that applies combined axial –sagittal bending loading through an axial preload along the knee-hip line and a transversal impact load at the mid-shaft site along anterior-posterior or posterior-anterior directions. The femoral tolerance curves calculated based on external loads show sensitivity with respect to the impact direction of transversal load due to the initial curvature of the femur, but insignificant dependence on the material mode, or the failure criteria used for femoral cortical bone. In addition to suggesting a numerical approach that uses finite element simulations and optimization techniques to determine the injury tolerance of long bones, the results highlight the predominant role of the bending loading in a combined loading of the femur.
Kelly S. Carney, Robert K. Goldberg, J. Michael Pereira – NASA Glenn Research Center, USA, Ryan S. Lee, Jeremie J. Albert – The Boeing Company
Reinforced Carbon-Carbon (RCC) is a ceramic matrix, coated C/C composite used in the Space Shuttle’s thermal protection system. A constitutive model for RCC is presented that separates the coating and substrate within a multi-material shell element allowing for bending strains to be accurately computed during a finite element analysis. Version 971 of LS-DYNA contains a new shell element formulation that allows for different constitutive models through the thickness. The carbon substrate is modeled with an orthotropic damage model that was originally developed by Matzenmiller, and has been enhanced into MAT_RATE_SENSITIVE_COMPOSITE _FABRIC (MAT 158). The behavior of the silicon carbide coating is modeled using a new constitutive model that is now included in LS-DYNA as MAT_NONLINEAR_ELASTIC_TENSION_CRACKS (MAT 236). The composite model is used for debris impact analysis on the RCC structures, and good correlation is shown with impact test deflections and damage.
Abdul Qadir Bhatti, Khaliq U Rehman Shad – National University of Sciences & Technology NUST, Pakistan, Norimitsu Kishi – Muroran Institute of Technology, Muroran
In order to establish a rational impact resistant design procedure for Arch type rock-shelter based on not only allowable stress design but also on ultimate state design and/or performance based design method, impact resistant capacity and/or maximum input impact energy for the RC structures must be clearly estimated. At present, the RC structures have been designed statically based on allowable stress design method. Here, maximum input impact energy for reaching ultimate state was numerically estimated by means of three-dimensional elasto-plastic finite element method for existing real arch-type RC rock-shelters with sand-cushion and EPS layer under falling heavy- weight impact loading. In this numerical analysis, solid elements were employed for concrete, falling heavy-weight and sand-cushion, and beam elements for rebar. Drucker-Prager and rebar yield criteria were used as material constitutive law for concrete and rebar, respectively. Cracks were estimated by allowing tensile stress cut-off at reaching at the tensile strength. In this paper, weight impact force, total axial force at the side-walls, displacement wave at the loading point, and crack patterns of the shelter at the time occurring of the maximum displacement are output. The results obtained from this study are as follows; 1). About twice the bending moment are generated at the edge as compare to the centre of falling weight in the direction of road axis. 2) It was observed that maximum response generation time of the bending moment and shear force is different. 3) When three layer buffer structure is set up in the tunnel pithead part, the sectional force can be decreased to about 1/2 to 1/5 in the mid-span compared with the case for sand cushion. Keywords: Pithead of tunnel; Three-layered absorbing system; Transmitted impact force; elasto-plastic response
Yang Hu, Changqing Du and Shyam Kariat – Chrysler, LLC, Li Zhang – Livermore Software Technology Corporation
M. P. Rao, M. Keefe – University of Delaware, Newark, B. M. Powers, T. A. Bogetti – United States Army Research Laboratory, Aberdeen Proving Ground
The objective of this study is to develop and demonstrate the feasibility of an LS-DYNA® Global/Local model for studying the non-linear mechanics of woven fabrics under ballistic impact. The presented approach is built on the observed response of fabrics in experimental studies performed at the Army Research Laboratory (ARL). In particular, two-layer test patches of 600 denier Kevlar KM2® fabrics are modeled with the aim of corroborating experimentally determined V50 velocities and physical deformation patterns. However, the present study begins with a brief overview of detailed three-dimensional (3D) finite element models of the woven fabrics under ballistic impact, comprised of regular undulating geometries of the individual yarn. This model is designated as the Full- Local environment and serves as the baseline for the subsequent Global/Local 3D finite element models. Within this Full-Local environment, the projectile velocity is determined as a function of time, and the response of the fabrics under the applied impact load are presented and discussed. Based on this work, the Global/Local modeling framework is developed that represents the fabric finite element meshes as comprised of combinations of homogenized continuum regions (‘Global’ regions) which neglect yarn undulations, and full 3D undulating yarns (‘Local’ regions). Discussions are presented regarding the implications on the predicted ballistic response of the fabrics. Specifically, comparisons are made for the predicted projectile velocity as a function of time, fabric deformations, energy histories, and computing time required to execute the individual simulations. It is shown that the Global/Local modeling approach results in reasonable savings in computing time without appreciably sacrificing the physics of the problem.
S. S. Akarca, W. J. Altenhof, A.T. Alpas – University of Windsor
The smoothed-particle hydrodynamics (SPH) technique was used to model experimentally observed large deformation behaviour of aluminum (1100 Al) during machining. The effectiveness of the SPH method in predicting the response of the 1100 Al workpiece during orthogonal machining has been assessed through a careful comparison with the experimentally measured stress/strain distribution within the chips formed during steady state cutting. An Eulerian numerical model, previously validated for machining of copper, was also used to evaluate the SPH model. Both the Eulerian and SPH models showed good overall correlation with the experimentally measured stress/strain distribution when an exponential stress-strain behaviour was utilized in modeling. The maximum predicted plastic strains utilizing an Eulerian and SPH solution approach were 7.5 and 8.0, respectively. When the CPU time requirements were considered, the SPH model was the suitable choice to model deformation processes during cutting with relatively good accuracy and approximately 2.75 times less cost compared to the Eulerian model.
Tatsuo Sakakibara, Toru Tsuda, Ryo Ohtagaki – ITOCHU Techno-Solutions Corporation
The SPH method is well known for the hydrodynamics and recently SPH is applied for the impact and the penetration behavior into solid materials. SPH is completely mesh free so that it seems to be appropriate to simulate crash behaviors of concrete and rock materials. However, the applicability and the effects of the parameter of SPH for quasi-static problems are not clear completely. In this paper, in order to verify the applicability and the accuracy of SPH on quasi-static problems, a number of unconfined uni-axial compression simulations are performed by SPH. The specimens of different number of elements are used and the parameter sensitivities of SPH are examined. The influences of formulation type of SPH in LS-DYNA® are also investigated. The results of SPH are compared with the Laglangian result. From the results, the effects of the parameter for SPH are made clear. Besides, it is found that the renormalized formulation is efficient to get the accurate results at boundary. As the result, it is conformed that SPH is applicable for the quasi-static problems.
Facundo Del Pin – Livermore Software Technology Corporation
The present work is an introduction to a new CFD solver in LS-DYNA®. This solver is part of the efforts put in LS- DYNA with the objective to expand the capabilities into new challenging problems in industry. The new CFD solver will focus on fluid-structure interaction (FSI) applications where incompressible fluids interact with the existing structures in LS-DYNA. Incompressible flows are present in a great variety of industrial applications from sloshing problems to aerodynamics at low Mach numbers. This new incompressible CFD solver coupled to LS-DYNA mechanics will provide an implicit time integration scheme allowing larger time steps and faster convergence to steady state. One of the main features of the solver is the Lagrangian representations of all FSI interfaces providing exact imposition of boundary conditions. In this way the fluid mesh and the solid mesh are tightly coupled such that the fluid domain deforms following the Lagrangian structure displacements. The rest of the domain follows an Arbitrary Lagrangian-Eulerian formulation. The image bellow shows the mesh movement and the conformity of the fluid mesh to the solid mesh.
Diane Xu – Ford Motor Company, Jim Kosek – Engineering Technology Associates, Inc.
One of the most critical issues in stamping manufacturing today is the successful shedding of scrap from limited trim dies. Until recently, die tryout was the first opportunity to check the shed scrap feasibility of a trim die. The newly developed scrap shed analytical module can be used for analyzing trim die scrap shed feasibility before die creation. It simulates Scrap shedding during or after the die is designed using Dynaform and LS-DYNA®. Scrap shedding simulation offers die designers and manufacturers the opportunity to closely examine a trim die’s performance before die construction. With today’s tighter die design timelines and reduced number of dies in manufacturing, it is more critical than ever to establish trim die design integrity as early as possible in the design process. This can be achieved through Dynaform scrap shedding simulation. Dynaform scrap shedding uses a flexible body approach to simulate the exiting of scrap from the workstation. This allows for full interaction of all essential variables and forces acting on the die and sheet metal part. It allows for a real world simulation that calculates the effect of any changes in die speed, initial velocity, material properties or die design. Various trim operations, such as direct and cam trim, can be very easily simulated. Once a design defect is found, possible solutions can also undergo a virtual tryout in the Scrap Shedding simulation. It has a great impact on cost and timing when used in stamping engineering, and can be used to avoid the pitfalls of defective die design.
M-A Lavoie, A. Gakwaya1 – Laval University, Cité Universitaire, Quebec, Canada, M. Nejad Ensan – National Research Council, Ottawa, Ont. Canada.
The SPH method is applied in industrial level problems as encountered for example in studies related to bird impact on composite aircraft structures. This paper first demonstrates the accuracy of the method for bird impact on rigid target modeling and then applies the developed model to a more complex problem, namely the secondary bird impact.
Guangye Li – IBM Deep Computing Group, Houston TX, Tushar Goel, Nielen Stander – Livermore Software Technology Corporation
The elitist non-dominated sorting genetic algorithm (NSGA-II) converges to the Pareto optimal front (POF) if a sufficient number of function evaluations are allowed. However, for expensive problems involving crash simulations, only a limited number of simulations might be affordable. It is observed that initially there are significant advances towards the POF but as the population matures, the improvements are relatively small. This means that one can probably limit the computational expense by terminating the search at the right point. The paper also demonstrates a successful use of IBM cluster for parallel processing that significantly reduces clock time for optimization.
Harihar Kulkarni, John M. Eidt – Ford Motor Company, Dharmveer Podhuturi, Akbar Farahani – Engineering Technology Associates, Inc.
Material handling of sheet-metal components within the plant or one plant to the other significantly impacts quality and assembly process. Any permanent deformation of sheet metal component contributes to poor quality of an automotive subassembly. One way is to minimize occurrence of unacceptable deformation is to follow trial and error approach. However, such approach is time consuming during launch; in addition cost of stamping and testing prototype panel is high. Therefore a CAE methodology, using dynamic time domain solver ‘LS-DYNA’, was developed to provide design guidance and to examine probability of permanent set in the panel due to manual and robotic handling loads. An outer panel of a liftgate is used as an example to establish this methodology. The CAE results confirmed that proposed design of outer panel is capable to meet handling loads and eliminated need of redesign and saved launch time. This paper discusses a methodology to simulate handling process and evaluate behavior of sheet metal panels subjected to time varying enforced displacements. In this approach FEA techniques and a nonlinear flexible dynamic model are combined. Such approach helped to simulate handling process and to assess relations among material strength, panel topology, and locations of suction cups in end-effectors. Experimental validation of this proposed technique is in progress.
Miles Thornton – Arup
Wenyu Lian – General Motors, Dilip Bhalsod – Livermore Software Technology Corporation, Lars Olovsson – IMPETUS Afea
The demands for developing safety restraint systems that perform well under Out-Of-Position (OOP) conditions have increased significantly in recent years. At the same time, the development of simulation capabilities for OOP have made progress in most major crash/safety software, such as the coupled Lagrangian-Eulerian approach (here referred to as AIRBAG_ALE) in LS-DYNA® by LSTC. Similar technologies are applied in MSC-Dytran by MSC and Madymo_CFD (Madymo) by TNO. A somewhat different approach is the FPM method in PAMCRASH by ESI. The AIRBAG_ALE capability in LS-DYNA, MSC-Dytran, and Madymo_CFD use loose-coupling techniques to couple the Lagrangian finite element airbag with a flow domain modeled with an Eulerian or ALE description of motion. PAMCRASH uses a particle based Lagrangian method, referred to as the Finite Point Method (FPM), for the description of the gases inside the airbag. All these methods share the same challenges associated with the coupling of the gas flow to folded bags under high speed deployment. Generally, the computations require considerable CPU power. The improved AIRBAG_ALE algorithm was developed in 2004 by Lian, Olovsson and Bhalsod . In the same year, a set of benchmark problems were proposed by Lian at the SAE Conference . A driver side airbag OOP study using AIRABG_ALE was presented in 2004 at LS-DYNA users’ Conference . During the last few years, some modeling difficulties using AIRBAG_ALE have been reported. To overcome the difficulties, the Corpuscular Method (here referred to as AIRBAG_PARTICLE) was developed by Olovsson . This study is intended as an evaluation of the accuracy, stability and efficiency of AIRBAG_PARTICLE compared to AIRBAG_ALE in OOP applications. More specifically, in this work the benchmark problems in ref.  have been studied using AIRBAG_PARTICLE. The results of AIRBAG_PARTICLE and AIRBAG_ALE are discussed.
José Imbert, Michael Worswick – University of Waterloo, Canada, Pierre L’eplattenie – LSTC, Livermore, USA
Electromagnetic Forming (EMF) is a high speed metal forming process that is being studied with interest in both academia and industry as a way of improving existing sheet metal forming. The main thrust of the published research has been increasing the formability of aluminum alloys. Observing and measuring EMF process is made very difficult by the high speeds involved and the tooling used to obtain the final shapes. As with many other processes, numerical simulations can potentially be used to study the details of EMF; however, one limiting factor is the difficulty in modeling the process, since accurate models of both the structural and electromagnetic phenomena must be solved. Researchers have relied on simplifications that use analytical magnetic force distributions, on separate electromagnetic (EM) and structural codes or on in-house codes that can solve both problems simultaneously. In this paper, the predictive ability of the EM module of LS-DYNA® is assessed through a comparison between experimental and numerical results for samples formed by EMF. V-Channel and conical shaped samples were formed using two different EMF apparatuses. For the V-Channel samples a double rectangular coil was used and for the conical samples a spiral coil was used. Both processes were modeled using LS-DYNA®’s EM module. A comparison of the final experimental and numerical final shape and current profiles is presented. It was found that the models provide good qualitative results and are able to predict the major features of the samples studied. Good quantitative agreement between the predicted and measured current profiles was found. The discrepancies observed are the result of numerical issues and the simplifications used in the creation of the specific models. It was found that the software can accurately predict the trends of the forming processes studied.
Karen E. Jackson, Yvonne T. Fuchs – NASA Langley Research Center Hampton
Simulation of multi-terrain impact has been identified as an important research area for improved prediction of rotorcraft crashworthiness within the NASA Subsonic Rotary Wing Aeronautics Program on Rotorcraft Crashworthiness. As part of this effort, two vertical drop tests were conducted of a 5-ft-diameter composite fuselage section into water. For the first test, the fuselage section was impacted in a baseline configuration without energy absorbers. For the second test, the fuselage section was retrofitted with a composite honeycomb energy absorber. Both tests were conducted at a nominal velocity of 25-ft/s. A detailed finite element model was developed to represent each test article and water impact was simulated using both Arbitrary Lagrangian Eulerian (ALE) and Smooth Particle Hydrodynamics (SPH) approaches in LS-DYNA®, a nonlinear, explicit transient dynamic finite element code. Analytical predictions were correlated with experimental data for both test configurations. In addition, studies were performed to evaluate the influence of mesh density on test-analysis correlation.
Abhijit Mahapatra, Avik Chatterjee – Central Mechanical Engineering Research Institute, India
This paper deals with the study of natural frequencies of vibration of continuous beams supported on hinged end supports with and without overhang. The paper illustrates the analytical formulation of the natural frequencies and corresponding modes of a n-span continuous beam by using a general solution for the Euler-Bernoulli differential equation. Using two different approaches, namely the analytical method and the numerical method some typical continuous beams are analyzed and the conformance of the FEM solver LS-DYNA is tested. The objective is to test the correlation between approximated analytical and numerical methods adopted for this particular study. The computed results are given in tabular form.
Stephen Kang, Paul Xiao – Ford Motor Company
Hybrid III Dummy Computer Aided Engineering (CAE) models have long been used for aspects of vehicle design, including vehicle structure and restraint systems to meet regulatory safety rating targets and internal company requirements. The quality and run time of these CAE models in simulating physical dummies directly affects the usefulness of CAE tools in vehicle development. The objective of this study is to compare the responses of the Madymo Rigid Body dummy and the LSTC Rigid-FE dummy in four crash test modes of one vehicle. During the study, in order to have a fair and sound comparison, the authors have requested each of the companies (TASS and LSTC) to examine the performance of their respective dummy models in essentially the same vehicle environment model. The authors would like to acknowledge their efforts and comments. In particular, frontal crash test modes using the passenger belted and unbelted 50th percentile male Hybrid III dummy and the passenger belted and unbelted 5th percentile female Hybrid III dummy were studied. In each test mode, the Madymo Rigid Body dummies and the LSTC Rigid- FE dummies (beta release) were used. The CAE model results were compared with test results in both time history measurements (acceleration, velocity, displacement, forces, moments) and occupant kinematics (using high speed video). This is the result of one case study and the authors do not intend to draw any general conclusions as to which dummy model is better in relation to the other.
A. Haufe, D. Lorenz – DYNAmore GmbH, Germany, K. Roll, P. Bogon – Daimler AG, Germany
In recent years the development of more and more niche products, i.e. cars with different external appearance, has become a remarkable trend in the automotive industry. This trend, however, generates higher costs for individual tooling geometries that are traditionally made as stiff as possible. A second trend is the increasing use of high and ultrahigh strength steel grades for bodies in white. Here too the design philosophy for the tools in sheet metal forming is based on a rather rigid and stiff tool approach. It is clear though, that a tremendous amount of money could be saved by designing the tools such, that their elastic deformation during the forming process is taken into account. This would lead to lighter and hence more inexpensive tools. The traditional approach to design the tool geometry by finite element simulations with rigid tools. Clearly, if elastic deformations are to be accounted for in such models, the assumption of a rigid tooling geometry needs to be abandoned. Here the straight forward approach would be to discretize the tool by a sufficiently accurate full 3D finite element model. Additionally the machine stiffness may be added to the model for completeness. Obviously this will lead to prohibitively increased computing time especially for large parts. A simple yet effective way to take the elastic deformations nonetheless into account is to condensate the discretized machine and tool geometry once and reuse it in subsequent simulations runs. The paper will discuss recent features in LS-DYNA® that allow the static condensation of elastic tool and machine geometries. Furthermore the application of the “deformable rigid bodies”- approach is shortly discussed.
Yuli Huang, Andreas Schellenberg, Stephen A. Mahin, Gregory L. Fenves – University of California at Berkeley
An adapter element provides a versatile and computationally efficient method for coupling several finite element (FE) analysis programs so that the unique modeling and analysis capabilities of each can be utilized simultaneously in the simulation of a complete system. FE software is coupled by (1) using each software’s programming interface to embed an adaptor element, and (2) connecting the adapter elements using OpenFresco (Open-source Framework for Experimental Setup and Control). The theory underlying the adapter element is based on the penalty method and communication of information at the nodes of the adapter element connecting a master with a slave program. The implementation and accuracy are then demonstrated using a dynamic analysis of a structural model from earthquake engineering.
Benjamin Tutt – Airborne Systems, R. Keith Johnson, Karen Lyle – NASA Langley Research Center
Airborne Systems (formally Irvin Aerospace Inc) has developed an Airbag Landing System design for the Orion Crew Module of the Crew Exploration Vehicle. This work is in support of the NASA Langley Research Center Landing System Advanced Development Project. Orion is part of the Constellation Program to send human explorers back to the moon, and then onwards to Mars and other destinations in the Solar System. A component of the Vision for Space Exploration, Orion is being developed to also enable access to space following the retirement of the Space Shuttle in the next decade. This paper documents the development of a conceptual design, fabrication of prototype assemblies, component level testing and two generations of airbag landing system testing. The airbag system has been designed and analyzed using the transient dynamic finite element code LS-DYNA®. The landing system consists of six airbag assemblies; each assembly contains a primary impact venting airbag and a non-venting anti-bottoming airbag. The anti- bottoming airbag provides ground clearance following the initial impact attenuation sequence. Incorporated into each primary impact airbag is an active vent that allows the entrapped gas to exit the control volume. The size of the vent is tailored to control the flow-rate of the exiting gas. An internal shaping structure is utilized to control the shape of the primary or main airbags prior to ground impact; this significantly improves stroke efficiency and performance.
Satoshi Fukushima, Koushi Kumagai, Tsuyoshi Yasuki – Toyota Motor Corporation, Yoshiharu Sonoda, Yu Setoyama – Toyota Technical Development Corporation
The use of occupant protection simulation by finite element model is increasing for vehicle development. Finite element models of moving deformable barrier honeycomb, vehicle body, dummy, and restraint system have been developed to predict the dummy injury values. To more accurately predict the dummy injury indices in side impact analysis, this paper describes the improvement of finite element models of restraint system such as torso airbag and polyurethane pad. The improved model of polyurethane pad was able to simulate both of anisotropy and breaking. The improved model of torso airbag was able to simulate the deployment behavior from inside of seat back foam. Finally, full vehicle level FE analysis was conducted to confirm the prediction accuracy of the dummy injury indices using the improved finite element models. The results show that the dummy injury indices correlate well with actual crash testing.
Tushar Goel, Nielen Stander – Livermore Software Technology Corporation, Yih-Yih Lin – HP
Genetic algorithms typically require a large number of simulations, which would be economically prohibitive for crash simulations without the advent of today’s cost-effective multi-core computers. A study is conducted to seek improvements while restricting the number of simulations and exploiting the ability to use parallelization. The parallelization, achieved by simultaneously running multiple simulations for each GA generation on a HP quad- core cluster, resulted in a significant time savings. Furthermore, the optimal distribution of computational effort to achieve the greatest improvement in performance was explored. A crashworthiness simulation of a vehicle with 58,000 element finite element model was used as a test example. Various population sizes and numbers of generations were tried while keeping the total number of simulations constant. The optimization performance is also compared with Monte-Carlo and space filling sampling methods. It is observed that using GA, one can find many feasible and trade-off solutions. It is beneficial to allow a greater number of generations to get good trade-off solutions. Significant improvements in the performance were observed.
Suresh Babu, Jayanta Biswas – Atomic Energy of Canada Limited
The present paper describes the drop analyses performed on a waste transfer flask using LS-DYNA® finite element software. Radioactive waste generated during retubing of CANDU®6 reactors are placed in waste containers. Waste transfer flasks are used to transfer waste containers from the reactor building to storage structures. The waste transfer flask is a complex mechanical assembly made of double wall steel shells filled with lead. At the storage area, the waste transfer flask is lifted by a crane to transfer the waste container to the storage structure. During lifting of the waste transfer flask, it is necessary to know the consequence of an unlikely drop event. Analyses have been performed to assess the structural response of the waste transfer flask subjected to impact loads due to various drop events resulting from lifting of the waste transfer flask at the storage area. The waste transfer flask is dropped by considering various orientations viz., dropping on edges, corners and base. Based on the drop analysis the mode of failure and plastic deformation of various components of the waste transfer flask and the waste container is predicted. Also, an assessment is made regarding the structural integrity of the waste container and its retrievability after various drop events.
R.E. Dick, J.W. Yoon -Alcoa Technical Center, USA
Wrinkling of thin sheet metal products such as beverage cans continues to be experienced by can manufacturers and is observed with some aluminum material suppliers more than others. These wrinkles are caused by compressive instabilities during forming, and with small base diameter cans and light gauge material, the likelihood of this wrinkle formation increases. Manufacturing experience suggests that dome wrinkling is influenced by many factors such as mechanical properties of the aluminum sheet, tooling geometry, contact conditions including the effects of lubrication, and other process boundary conditions. It is also difficult to conduct an experimental analysis of compressive instabilities of these thin sheet metal products because the effects of all of the factors contributing to the instabilities are complex and small changes in these factors may produce widely varying results. Therefore, a numerical approach is recommended to separate the effect of each variable on wrinkle formation. This paper shows how strain-hardening and r-values influence wrinkle formation in its magnitude and frequency through dome forming of a beverage can based on a recent anisotropic yield function implemented as an LS-DYNA UMAT subroutine.
Dr.-Ing. Matthias Hörmann – CADFEM GmbH, Germany
Ken-An Lou, William Perciballi – ArmorWorks
The objective of this project was to develop innovative lightweight mine-protected fasteners for blast protection appliqués. Blast protection appliqués are used on tactical and combat ground vehicles as a method of deflecting or mitigating the effects of anti-vehicular mine blasts or attack by Improvised Explosive Devices (IEDs). A critical and weakest component of these appliqués is the fastener joints. Currently, industrial bolts are commonly used for attaching blast protection appliqués to vehicles due to simplicity. However, under blast conditions, these bolts can often shear off causing secondary fragments and projectiles which may impact the vehicle and crew, causing additional damage and injury to the vehicle personnel. This study focused on conducting three-point bending tests to evaluate different bolt materials. Bolt preload or stress initialization was simulated via LS-DYNA® Implicit with two separate analyses. Also a similar two-step initial strain analysis was performed using NEiNastran. The simulated results will be compared with test data. Future work will include finite element modeling and testing of fasteners under dynamic, ballistic, and blast loads.
Geneviève Toussaint, Robert Durocher – Defence Research and Development Canada – Valcartier
Light Armoured Vehicles (LAV) are essential to transport troops and equipment in combat and high risk zones. The emergence of non conventional threats such as Improvised Explosive Devices (IED) implies a need to reassess armoured vehicle design to improve crew survivability. To do this DRDC has been developing new experimental facilities and has conducted experimental testing to study armoured vehicle structure subjected to near field blast loadings. Numerical models of the test facility were developed to assist the design of the experimental program and to accelerate the design of improved protection systems. This paper presents results of experimental and numerical analysis, in order to compare the numerical predictions with experimental results. In this study the smoothed particle hydrodynamics (SPH) technique was applied to model the loading on the vehicle structure.
Hongsheng Lu, C. T. Wu, Jingxiao Xu – Livermore Software Technology Corporation
With the recent improvements in adaptive procedure, adaptive mesh-free method has become an important tool to solve the 3D forging and extrusion problems that usually involve large topology change with severe local deformation. In this paper, the implicit version of 3D adaptive EFG method with emphases on the state variable transfer between successive discretizations is presented, and several problems are used to illustrate the effectiveness of the proposed approach. A massively parallel processing (MPP) of adaptive EFG is implemented for the 3-D analysis, and the scalability test is conducted on the highly deformable inelastic example problems.
Bill Feng, David Coleman – Jaguar Cars and Land Rover
The curtain airbag is usually simulated by a uniform pressure method in which the pressure of the airbag is considered as constant. This is correct when the airbag is fully deployed. However, this assumption is not valid during the curtain airbag deployment phase, where the gas flow passing through each chamber can be clearly seen. Therefore it is impossible to simulate the intermediate sequences of the airbag deployment correctly by using the conventional uniform pressure method. The gas dynamic module has been made available within LS-DYNA®. This module provides the basic toolset to simulate gas flow. This technique enables us to simulate the gas flow and pressure distributions in detail and enables the correct sequences of the curtain airbag deployment. Interactions between the airbag and interior trims can now be simulated and well understood. This gas dynamic simulation method can be implemented to identify the potential failure modes of the curtain airbag during deployment through trims in the design development stage of the programme. Therefore, the confidence level for “right first time” curtain airbag deployment can be greatly increased.
Uli Franz, Alexander Gromer, Jens Zschieschack – DYNAmore GmbH, Matthias Walz – Mercedes Car Group, Daimler AG, Yupeng Huang – Sunrise Auto Safety Technology
Very fine models allow investigating the behavior of dummies with high accuracy. Even if ele- ment numbers for a dummy model above 3 Million elements are currently not suitable for stan- dard simulation in vehicle development, the usage of such models contribute to the development of coarser dummy models. Due to the detailed representation physical effects occurring can be captured very realistically with the very fine models. The paper present current methodology to develop models within the FAT or PDB frame work and outlines first experiences with very fine models to enhance coarse dummy models.
Venkatapathi Tarigopula, Magnus Langseth, Odd Sture Hopperstad – Norwegian University of Science and Technology
This paper presents a systematic numerical investigation of the influence of element formulation on the force- deformation characteristics and crush behaviour of thin-walled dual-phase steel square tubes subjected to axial loading. Influence of shell and volume elements were verified on the crush behaviour. Finite element models square sections were created and analysed using the non-linear explicit finite element code LS-DYNA®. Parameters of interest were the energy absorption, peak crush load capacity, and the crush behaviour. The strain-rate effect has been considered for the dynamic simulations. Even though the initial peak load from numerical analyses differs significantly from the tests, the mean force does not deviate greatly whatever may be the element formulation. Both shell and brick elements predicted the experimental responses reasonably well. However, the element aspect ratio in volume element simulations seems to play a role in accurately capturing the local bending response when subjected to axial compressive load.
Tushar Goel, Nielen Stander – Livermore Software Technology Corporation
Meta-models are frequently used to offset high computational cost of crashworthiness optimization problems. Radial basis function based meta-models are gaining popularity among various meta-modeling techniques due to their ability to approximate non-linear responses with relatively low fitting cost. However, the performance of RBF networks is very sensitive to the choice of topology. In this paper, the influence of three selection criteria namely, PRESS, pointwise PRESS error ratio, and estimated variance of noise, over network topology is studied. The results are demonstrated for a few analytical functions and a crashworthiness simulation of a full NHTSA vehicle problem. The results showed that the PRESS-based method was the most reliable method to select network topology.
Ibai Ulacia, Iñaki Hurtado – University of Mondragon, José Imbert, Michael J. Worswic – University of Waterloo, Canada, Pierre L’Eplattenier – Livermore Software Technology Corporation, USA
In this paper the Electromagnetic Forming process is numerically analyzed using a novel method that is being implemented into the commercial LS-DYNA® code. The method consists on a combination of a Boundary Element Method (BEM) and a Finite Element Method (FEM) formulations to compute the electromagnetic analysis. The main advantage of using BEM is that the air is not meshed avoiding the related problems and making it easier to couple the interacting fields involved in the EMF process. Sequential coupling and uncoupled strategies for the simulation of the related fields (i.e. electromagnetic, thermal and mechanical fields) are compared. Sequential coupling takes into account the deformation of the work piece are compared with the uncoupled strategy dismisses the deformation of the sample. The importance of the simulation method used is evaluated and compared with experimental results.
S.V. Medvedev, M.V. Petrushina, O.P. Tchij – United Institute of Informatics Problems NAS of Belarus
Welded elements of a two meter long tube is considered. The welded structure rotates with 100rad/sec. The non- symmetrical brackets are connected to the tube through by the welds of a curved trajectory that consists of circumferential parts and arbitrary curved part on a shaft surface and disposed on the tube in a helical sequence. The objective of the study was to explore the axis and hoop stress distribution on the shaft tube and its influence on the dynamic properties of a welded structure. Sample with one element welded to the shaft tube first was studied. Three approaches were used. Temporal and residual stresses and strains were first obtained by means of moving heat source and by means of solving simplified thermomechanical problem assuming weld laying on being simultaneous along the weld path. Cooling process was important in both methods. Third approach is an own method based on shrinkage forces notion which implementation needs only strength analysis formulation. Its correctness and applicability to the case of weld paths under study was checked. The residual welding deformations obtained by all methods were then compared with the deformations of samples welded in manufacturing conditions.
Tim Prince – Intel Corporation
The Intel Cluster Ready program enables LS DYNA/MPP users to buy, install, and use clusters more effectively. It includes a joint Intel and cluster supplier certification process to ensure the cluster the LS-DYNA user purchases is designed and built to specification. Intel supplied software tools support verification of initial and ongoing operation and performance of the cluster.
Olivier Schreiber, Michael Raymond, Srinivas Kodiyalam – SGI
Multi-Rail networks can improve MPI communication performance by distributing the communication traffic to multiple independent networks (rails). Messages are divided into several chunks and sent out simultaneously using multiple rails. With the dual plane network topology of SGI Altix ICE clusters, MPI communication can hence utilize both the InfiniBand rails, including, ib0 and ib1 fabrics. The performance gains achievable with LS-DYNA for complex crashworthiness simulations through the use of MPT dual-rail over MPT singe-rail on an Altix ICE system are indeed significant.
I. Makropoulou, Y. Kolokythas, L. Rorris – BETA CAE Systems S.A., Greece
In the presently CAE-driven vehicle design process a great number of discipline models must be built and analyzed for the validation of a new vehicle model design. The increasing number of vehicle model variants further increases the number of the load-cases that must be studied. This process introduces a great amount of disparate data that need to be handled by the CAE teams. However, due to the multiple sources and the diversity of the CAE data, the current level of organization and data management deployed does not account for them. Setting as a target the reduction of the CAE turnaround cycle and cost, the pre-processing tools are required to streamline all “input” data and at the same time the simulation model build-up process itself, .this paper will present the means provided by BETA CAE Systems S.A. towards the development of realistic, repeatable and robust crash simulation models for LS-DYNA. ANSA Task Manager, using template processes, supervises the generation of the simulation models, while ANSA Data Management, in the background, facilitates the components management, ensuring that the engineering teams will always work with the most up-to-date data. The simulation model set-up becomes a repeatable and user-independent procedure, safeguarding the model quality and fidelity
L. J. Deka, U. K. Vaidya – The University of Alabama at Birmingham
The impact damage response of balsa core sandwich composite plates with S2-glass/ epoxy reinforced facesheets is evaluated by impacting them with a spherical steel projectile at single impact locations. The impact damage can significantly reduce the structural integrity and load bearing capacity of a composite structure. Under high velocity impact loading, laminated composites experience significant damage causing fiber breakage, matrix cracking and delamination. Energy absorption and delaminations from high velocity impacts with spherical projectile of .30 caliber is discussed. Finite element modeling was used to gain insight into failure modes, energy absorption, and damage prediction. During high velocity impact, composite laminates undergo progressive damage and hence, Material Model 162, a progressive failure model based on Hashin’s criteria, has been assigned to predict failure of the laminates. The laminates, the projectiles and the balsa wood core are meshed using brick elements with single integration points. These results were then compared with experimental data obtained from three layer S-2 glass/epoxy facesheets balsa core sandwich structures. An excellent correlation between experimental and numerical results had been established.
Ismael Henchi, Pierre L’eplattenier, Nielen Stander – LSTC, Glenn Daehn, Yuan Zhang, Anupam Vivek – The Ohio State University
In this paper, a parameter identification procedure to obtain the constitutive properties of metals at high strain rate and high temperature is presented. This procedure uses experimental results from electromagnetic ring expansions, coupled with LS-DYNA® simulations using the newly developed electromagnetism module, driven by LS-OPT®. The experiments were performed at The Ohio State University and the expansion velocities of the ring were measured. These are used as the target data for the optimization process where the constitutive properties are varied. The procedure is presented in details. It is then tested on a numerical case where the target velocity was generated by a simulation with given constitutive properties. Finally, it is used to find the constitutive properties of a copper alloy.
Roger G. Grimes, Xinhai Zhu – Livermore Software Technology Corporation
The authors will present the use of LS-DYNA for a variety of metal forming applications. They will present some new features and improvements in Version 971 of LS-DYNA such as Inertia Relief and Contact Penetration Detection. The presentation will include applications of gravity loading, binder wrap, flanging, springback and die transfer.
Nielen Stander, Tushar Goel – Livermore Software Technology Corporation
A study is conducted to determine the sensitivity of 2 topologically distinct metamodel types to variations in the experimental design brought about by sequential adaptive sampling strategies. The study focuses on examples encountered in crashworthiness design. Three sampling strategies are considered for updating the experimental designs, namely (i) a single stage approach, (ii) a sequential approach and (iii) a sequential approach, but with higher densities in local regions. The experimental design type is the Space Filling Method based on maximizing the minimum distance between any two design points within a subdomain. Feedforward Neural Networks (NN) and Radial Basis Function Networks (RBF) are compared with respect to their sensitivity when applied to these strategies. A large set of independent checkpoints, constructed using a Latin Hypercube Sampling method is used to evaluate the accuracy of the various strategies. Four examples are used in the evaluation, namely (i) simple two- variable two-bar truss, (ii) the 21 variable Svanberg problem, (iii) a 7 variable full vehicle crash example and (iv) a 11 variable knee impact crash example. The example, analyzed using LS-OPT® for metamodeling and LS-DYNA® for FE modeling, reveal the following: while expensive to construct, NN committees tend to be superior in predictability whereas the much cheaper RBF networks, can sometimes be highly sensitive to irregularity of experimental designs caused by subdomain updating. However, this conclusion cannot be extended to the crash problems tested, since the RBF networks performed consistently well for these examples.
Akram Abu-Odeh – Texas Transportation Institute
Bridge rails are constructed to contain and redirect an errant vehicle. They are constructed to withstand the impact severity of such vehicular impact based on the desired containment level. To evaluate the structural integrity of a given bridge rail design, bogie tests were conducted using a 5000 lb bogie as an impactor. In this paper, LS-DYNA was used to model the concrete barrier to simulate the bogie impact. Three material models in LS-DYNA (type 72R3, 84 and 159) were used to simulate the impact event. The rebars to concrete coupling was modeling via the *CONSTRAINED_LAGRANGE_IN_SOLID feature in LS-DYNA. Time history and deformation profile comparisons between tests and LS-DYNA simulations are presented in this paper. Figure 1 below shows a damage profile for the barrier as tested and as simulated in LS-DYNA.
C. Espinosa, M. Salaun,C. Mabru,R. Chieragatti – Université de Toulouse, France, J.L. Lacome – LSTC, USA, J. Limido IMPETUS Afea, France
The purpose of this work is to evaluate the use of the Smoothed Particle Hydrodynamics (SPH) Method within the framework of modeling high speed cutting. First, a 2D SPH based model is carried out using the LS-DYNA® software. SPH is a meshless method, thus large material distortions that occur in the cutting problem are easily managed and SPH contact control allows a “natural” work piece/chip separation. The developed SPH model proves its ability to account for continuous and shear localized chip formation and also correctly estimates the cutting forces, as illustrated in some orthogonal cutting examples. Then the SPH model is used in order to improve the general understanding of machining with worn tools. At last, a milling model allowing the calculation of the 3D cutting forces is presented. The interest of the suggested approach is to be freed from classically needed machining tests: Those are replaced by 2D numerical tests using the SPH model. The developed approach proved its ability to model the 3D cutting forces in ball end milling.
John D. Reid, Robert W. Bielenberg – University of Nebraska-Lincoln
Modeling energy absorbing foams that restore back to their original shape can be extremely challenging, especially when the foam is crushed over 90%. However, the foam used in the SAFER racetrack barrier does indeed nearly return to its’ original shape after severe crushing. The desire to model this behavior led to the use of the Fu Chang foam model available in LS-DYNA .
P.K.C. Wood, C. A. Schley – University of Warwick, England, M. Buckley – Jaguar and Land Rover, England, B. Walker, Ove – Arup and Partners, England, T. Dutton – Dutton Simulation, England
This paper describes a new model and method to predict Self-Piercing Riveted (SPR) joint interlock failures in aluminium sheet at crash speeds using explicit finite element simulation. SPR interlock failure is dependent on rivet direction, which is included in the model. A mesh independent approach is adopted for connection model which is capable of industrial application at the full vehicle crash analysis level. The paper provides an overview of the approach to validate connection model; typically by developing detailed physics based models of various joint configurations supported with high speed experimental data, through to model capable of industrial application. The framework to validate joint failure model for use in crash simulation tools is expected to have broader application.
Thomas Borrvall – Engineering Research Nordic AB, Sweden
A challenging task for the static implicit nonlinear solver in LS-DYNA is to accurately and robustly solve contact problems, especially is this needed for stamping simulations. This paper aims at investigating the benefits of a mortar segment-to-segment contact algorithm by Puso and Laursen [1,2] when compared to the traditional node-to- segment approach. A penalty based version of the algorithm is implemented in LS-DYNA, meaning that the contact tractions are proportional to both the penetration as well as the overlapped area of segments in contact. This allows for the nice property that the resulting global contact force is continuous with respect to deformation and thus makes the approach intuitively suitable for implicit analyses. Further measures for smoothing the response are implemented in the method and the first tests indicate that the method is advantageous at least for a certain class of problems, but how great the overall impact will be remains to be seen.
Gaëtan Van den Bergh, Yves Lemmens – LMS International
Automotive companies have become increasingly educated in safety related performance, setting the bar higher for automotive crashworthiness. Process Automation and Design Optimization (PIDO) tools help LS-DYNA users to reach these safer designs in a shorter period. In this paper, various application cases are used to demonstrate how PIDO tools can leverage your existing simulation codes. In the first application case, a multi-disciplinary crash optimization case, executed at an automotive company, is discussed in detail. How can you optimize systems while taking into account multiple different government safety regulations? How can you reduce the throughput time of your CPU intensive LS-DYNA simulations? How can you manage and visualize the obtained data? During the second application case, an industrial example of an A-pillar trim design optimization for occupant head safety is presented. During the optimization the variability present in actual testing conditions is also taken into account. The analysis aims mainly at minimizing head injury by using a Reliability-based approach that takes into account a probabilistic constraint formulation
Mehrdad Asadi – Cellbond Composites Ltd., UK, Brian Walker – Arup, Hassan Shirvani – Anglia Ruskin University, UK
The US Federal Standard for Side Impact Protection (FMVSS 214) uses a deformable barrier and defines the dimensions and materials of the barrier, as well as the crush strength of the aluminium honeycomb parts in the main block and the bumper. This deformable barrier is also used for rear impact according to the updated FMVSS 301. This paper represents the methodology to create the advanced Finite Element model of Cellbond’s NHTSA barrier and validation through experimental test data. The explicit LS-DYNA® was used to analyze the model while number of static compressive tests performed at different angles to characterize aluminum honeycomb Material Cards. A strain-rate scale factor curve is defined to simulate the dynamic stiffening in the aluminium honeycomb during the analysis. Adhesive properties are also obtained using Climbing Drum, T-Peel, Tensile and Plate Shear test results. The preliminary component tests generated a good correlation with FE outputs and to validate the barrier model, similar impact tests were performed in LS-DYNA environment respecting to three experiments Flat-Wall, Rigid-Pole and Rear-Armature tests. In all assessments, the barriers were mounted on a moving trolley and were tested at certain speeds. The Final comparison on overall results represents a good correlation between test data and CAE results for all tests.
Stan Posey – Panasas, Inc., Fremont, CA, USA
As HPC continues its aggressive platform migration from proprietary supercomputers and Unix servers to HPC clusters, expectations grow for clusters to meet the I/O demands of increasing fidelity in CAE modeling and data management in the CAE workflow. Cluster deployments have increased as organizations seek ways to cost- effectively grow compute resources for CAE applications, and during this migration many also implemented conventional network attached storage (NAS) architectures to simplify IT administration and further reduce costs. While legacy NAS implementations offer several advantages of shared file systems, most are too limited in scalability for effective management of I/O demands with parallel CAE applications. As such, a new storage migration is underway to replace legacy (serial) NAS with parallel NAS architectures and parallel file systems. This new class of parallel file system and shared storage technology was developed to scale I/O in order to extend the overall scalability of CAE simulations on clusters. This paper examines CAE motivation for shared parallel file systems and storage, for requirements of multi-physics LS-DYNA® applications on conventional clusters with proper balance for I/O. Model parameters such as size, element types, schemes of implicit and explicit (and coupled), and a variety of simulation conditions can produce a wide range of computational behavior and I/O data management demands. The benefits of a Panasas storage implementation are introduced for such broad requirements, through examples of CAE workflows for a variety of production-level applications in industry.
Gaurav Nilakantan, Michael Keefe, John W. Gillespie Jr. – University of Delaware, Travis A. Bogetti – US Army Research Laboratory
A novel approach to the multi scale modeling of the impact of woven fabrics using LS-DYNA® has been presented. This new technique entitled ‘Hybrid Element Analysis (HEA)’ incorporates the use of different finite elements at both a single and multiple level of modeling. A yarn level resolution is maintained around the impact zone or local region, while a homogenized resolution has been used for the far field or global region. The central patch of yarn level resolution uses a combination of solid and shell elements. A new method for modeling individual yarns using shell elements is discussed, which more accurately captures the geometrical contours of the yarn cross section. The surrounding homogenized zone uses shell elements. Interfaces using various types of tie-constraints are created between the different finite elements at the various scales of modeling. The acoustic impedances have been matched across the interfaces. A systematic approach is presented to determine the geometric and material parameters of the homogenized zone. The HEA approach maintains the accuracy of using a fabric model comprised entirely with yarn level resolution utilizing solid elements, but at a fraction of the computational expense. This enables the finite element simulation of multi layered fabric systems with very large domains, which was previously very difficult because of the impractical computational requirements of such an exceedingly large model. Compared to previous numerical multi-scale models, the finite element model using the HEA approach presented in this paper more accurately captures the entire impact event at a lower computational expense, making it a very useful tool for future studies.
Neil Turton, Amitabha Majumder, William Altenhof, Daniel Green, Vivek Vijayan, Honggang An,Shun Yi Jin – University of Windsor, Canada
An experimental and numerical investigation was conducted on AA6061-T6 extrusions with centrally located through-hole discontinuities to investigate the strains in the vicinity of the crush trigger. The extrusions used in this research were of square cross sectional geometry with a nominal side width of 38.1 mm, wall thickness of 3.15 mm and length of 200 mm. Centrally located circular discontinuities with a diameter of 14.29 mm were incorporated into the extrusion through CNC machining. The axial crushing tests were performed in a quasi-static manner using a hydraulic Tinius-Olsen tension/compression testing machine and observed using a GOM Aramis optical strain measuring device focusing on the region in the vicinity of the discontinuity. A finite element model previously developed by Arnold and Altenhof was selected to compare strains in the region of the through-hole. The material definition for the AA6061-T6 extrusion utilized in this FE model incorporated damage mechanics theory which was able to accurately predict material failure after a two-stage calibration process. All simulations considered in this research were completed using LS-DYNA® version 970 revision 5434a. Good predictive capabilities of the strain magnitudes and distributions were observed from the numerical simulations. In the vicinity of the discontinuity, at the stress concentration region and at maximum crushing force, the effective strain was observed to range from 4.5% to 15.5% in LS-DYNA simulations. Experimental observations on the effective strain ranged from 5.0% to 15.7%. This information is important in the understanding of large deformation behaviour for AA6061-T6 extrusions, which could be applied in structural crashworthiness applications.
Yuan Zhang, Geoff Taber, Anupam Vivek, Glenn Daehn, Pierre L’Eplattenier – Livermore Software Technology Corporation, Suresh Babu – The Ohio State University
Magnetic Pulse Welding (MPW) is a collision welding process, similar to Explosive/impact Welding (EXW), but it utilizes electromagnetic force as the acceleration mechanism. Therefore, the available energy is much lower than EXW and it makes the process safer and more reproducible for sheet seam welding. However, the available energy must be better focused and controlled. In the sheet seam MPW process, a flyer sheet is driven and collides with a target sheet. True metallic bonding is achieved at the mating interface if contact takes place above a critical impact velocity at an appropriate impact angle. The impact velocity and angle are determined by the primary and induced electromagnetic fields. Both of them are strongly related to the geometry of the electromagnetic actuator and the discharge characteristics. An MPW launch system that will robustly provide bonding can either be developed empirically or through simulation. Here we attempt to provide the basis for a simulation-based approach to system design. The oblique MPW impact of AA606-T61 and Cu101 were analyzed using the newly available LS-DYNA®. Electromagnetism (EM) module in This module allows performing coupled mechanical/thermal/electromagnetic simulations. The simulation can predict the impact velocities and the temperature distribution along the mating interface. The simulation results were validated with measurement by Photon Doppler Velocimetry (PDV) measurement. Additionally, the simulation results also indicated rapid thermal cycling on the mating interface.
F. Neukamm, M. Feucht, K. Roll – Daimler AG, Germany, A. Haufe – DYNAmore GmbH, Germany
With increasing requirements on crashworthiness, and light-weight car body structures being a central issue in future automotive development, the use of high strength steel qualities has become wide-spread in modern cars. Since these materials often show significantly lower ductility than conventional steels, it is of great importance to precisely predict failure under crash loading conditions. Hence constitutive models in crashworthiness applications – as for instance the Gurson/Johnson-Cook model which is applied widely at Daimler AG – need to be initialized with correctly determined internal variables mapped from a corresponding sheet metal forming simulation. Here two principle ways could be used theoretically: On the one hand different understanding of damage and failure in crashworthiness and sheet metal forming applications may be unified by a generalized incremental stress state dependent damage model (GISSMO). This approach can be considered as an attempt to replace the currently used FLD for the failure description in forming simulations. Furthermore, an advantage would be the inherent ability to account for load-path dependent failure behavior. On the other hand the already applied Gurson model in crash simulations may be fed by an estimation of the internal damage value from the forming simulation. The idea here would be to perform the forming simulation with a state-of-the-art anisotropic material model like e.g. the Barlat model, with a simultaneously executed estimation of Gurson’s damage evolution law. The present paper will enlighten these two possible approaches. Furthermore it will be shown that damage prediction in metal forming processes and subsequently the use of the results as initial damage values in crash simulations is possible and necessary to predict structural failure in crashworthiness simulations.
Honggang An, Daniel E. Green – University of Windsor, Canada
Optimization and sensitivity analysis is important although difficult to obtain for tubular hydroforming, because of the implicit relationship between the load path variables (internal pressure and tube end displacement) and the dependent variables (such as stress, strain and final tube thickness). In this paper, the Taguchi method was used in conjunction with virtual hydroforming experiments using LS-DYNA® to carry out the sensitivity analysis and optimization of straight-tube hydroforming. This method employs an orthogonal array to study a large parameter space using only a small number of numerical simulations. Since the tube wall undergoes bending as it fills the corner of the die during the final stage of hydroforming, the strain path becomes non-linear. In this situation, the traditional strain forming limit diagram (FLD) is not a reliable criterion for necking/fracture. In contrast, the forming limit stress diagram (FLSD) is practically strain path-independent. Therefore, the FLSD was adopted for the necking/fracture criterion of the process. Multi-objective functions that consider necking/fracture, wrinkling and severe thinning were taken to evaluate the performance of each simulation. The Pareto optimum was obtained with a minimum failure value using the minimum distance method. Furthermore, the analysis of variance (ANOVA) statistical method was used to determine the effects of the forming parameters on the quality of the final hydroformed part. The factor response was completed using the Signal-to-Noise (S/N) ratio and ANOVA results. The ANOVA indicates the degree of sensitivity for the hydroforming parameters, and expansion pressure, calibration pressure, and tube end displacement were shown to be the three most important factors. This combination of numerical analyses and an optimization technique has helped to define a load path that leads to a robust manufacturing process and good part quality. Keywords: Tube hydroforming, FLSD, Taguchi method, Sensitivity analysis, Pareto optimization
Gilad Shainer, Swati Kher – Mellanox Technologies
Increasing demand for computing power in scientific and engineering applications has spurred deployment of high- performance computing (HPC) clusters. Finite Element Analysis (FEA) and Computational Fluid Dynamics (CFD) are computational technologies that can take advantage of HPC clusters for increasing engineering design productivity, reduce development cost and faster time to market. The end-user benefits are far more sophisticated, enhanced, safer and robust products. With increase usage of multi-core in HPC clusters, FEA and CFD applications need to be highly parallel and scalable in order to fully utilize cluster computing ability. Moreover, multi-core based clusters impose higher demands on cluster components, in particular cluster interconnect. In this paper we investigate the optimum usage of high-performance clusters for maximum efficiency and productivity, for CAE applications, and for automotive design in particular.
Jaeho Shin, Costin D. Untaroiu, Jeff R. Crandall – University of Virginia Center for Applied Biomechanics
Since the limited space between the car structure and an occupant makes it difficult to manage side impact energy, much biomechanical investigation has been done by subjecting the pelvis to lateral loading. In this study, the dummy finite element model was partially modified to verify the lateral pelvic loading by a rectangular shape impactor and used to explain in detail the previous investigation under iso-energy. In order to better understand the influence of impact mass and velocity under iso-energy, linear momentum and total energy conservation theories were introduced. Using driven equations from the theories and the simplified pelvis model, this study proved that the maximum internal energy levels should be different under iso-energy: the greater the impact mass, the less the internal energy level. This finding correlates with the previous pelvis loading investigation: the greater the impact mass, the less the pelvic loading, since it was shown that the impact loading is proportional to the internal energy. Thus, this study calculated the impact mass and velocity combinations to maintain the same internal energy level based on analytical solutions and finite element simulations. Closed values of the maximum pelvic forces were obtained when the calculated impact mass and velocity conditions were applied on the dummy lateral impact model. Furthermore, this methodology in conjunction with the analytical solution and the finite element simulation should be an appropriate way to set up the impact test configurations using manageable internal energy levels which may help to better understand the loading characteristics.
David J. Benson – Dept. of MAE, UCSD, Yuri Bazilevs, Thomas J. R. Hughes – ICES, The U.T. Austin
Piecewise continuous Lagrangian polynomials are the traditional interpolation functions used in the finite element method. They work well for many applications, but they also have shortcomings for many important applications. For example, in metal forming, the dies are designed using CAD programs and their geometry is defined in terms of NURBS (non-uniform rational B-splines) which can not be exactly replicated with a piecewise continuous Lagrangian polynomial in all cases. Therefore, there is a geometric incompatibility between the desired shape and the kinematic range of the blank modeled with traditional finite elements. This paper presents initial results for a shell element formulation based on NURBS.
David J. Benson – Dept. of Structures, UCSD, Yuri Bazilevs, Thomas J. R. Hughes – ICES, U Texas Austin
Gaël Le Blanc,Patrice LaPorte,Gilles Avrillaud, Paul Vincent – ITHPP, France, Pierre-Yves Chanal, Pierre-Louis Hereil – Centre d’Etudes de Gramat, France, Pierre L’eplattenier – LSTC, USA
GEPI  is a pulsed power generator developed by ITHPP for Centre d’Etudes de Gramat (CEG) devoted to ramp wave (quasi isentropic) compression experiments in the 1 GPa to 100 GPa range and to non shocked high velocity flyer plate in the 0.1 km/s to 10 km/s range. The aim of this paper is to compare numerical simulation and experimental results on a 3D GEPI configuration. A coupled mechanical / thermal / electromagnetism simulation has been performed to validate the new LS-DYNA beta version, ls980, where the electromagnetism package has been implemented. The validation is realized by free surface velocity comparisons on a reference configuration . The study is presented in three main steps. First, a validation of LS-DYNA is performed by comparison to an analytical model and to another electromagnetic solver modeling a basic configuration. In the second step, the GEPI configuration is described and experimental strip line free surface velocities are analyzed. Then LS-DYNA models are presented. The model validation is realized by free surface velocity comparisons between LS-DYNA and GEPI results. As the electromagnetism solver requires a lot of memory, in order to optimize the memory used as well as the CPU time, the electromagnetic domain is limited to the “launcher” with a thickness around 1 mm. The rest of the strip line is merged to the “launcher” and the mechanical solver only is used there. The last section of this study presents the optimization of the strip line geometry in order to improve the magnetic pressure homogeneity along this strip line.
Matt Clarke – Continental Tool and Die, Jeanne He – Forming Simulation Technology LLC
In a modern day draw simulation; our objective has always been to verify the formability of the deformed blank. We then utilize the output of the simulation to ascertain the forces required to form the part. Little time is spent attempting to verify if our design for the die is capable of reproducing these results. Most simulation assumes the tools are rigid. The real expertise comes when you can reproduce that scenario in an actual tool that makes parts in a consistent manner. This study follows a real world die development and build project, where the initial tryout was completely different from the simulation results, binder deformation has played a key role which differs the simulation results in which all tools are assumed to be rigid. Further simulation with a flexible binder has been performed, also compared to the real world solutions that were developed to make a good part. This study also provides valuable information for exploring the next generation of forming simulation needs. A major advance in simulation technology would be to answer the question of how simulation can compensate for these inadequacies. Through this study, it is clear that optimization analysis for various tooling needs to be shortened the tooling process time and reduction of the cost is an obvious trend in the near future.
Leonard E. Schwer – Schwer Engineering & Consulting Services
Analysts faced with making predictions that involve uncharacterized materials need to bound their results in a rational manner. For concrete materials, LS-DYNA® includes three simple input concrete material models that provide strength envelopes ranging from low, to intermediate, and high strength. Using a range of concrete strengths in numerical simulations is rational when little or no other information about the concrete is available. The typical system response quantity of interest (SRQ) in perforation studies is the exit velocity of the projectile. The simple input concrete models provide a range of concrete resistance to penetration that in turn produces a corresponding range of projectile exit velocities. Quantifying this possible range of exit velocities provides the decision maker with an assessment of what to expect in the field or laboratory for uncharacterized concrete targets. The use of multiple simple input concrete models to provide a range of exit velocities is demonstrated for a simple model of a brick wall, where no experimental data was available. To provide the reader with some indication how this approach can be used to compare predictions with laboratory data, the experimental and numerical results for a set of concrete cylinder perforations is provided.
Yun Huang – Livermore Software Technology Corporation, Mhamed Souli – University of Lille, France
The present work concerns the new capability of LS-DYNA® in solving acoustic and vibro-acoustic problems. In vibro-acoustic problems, which are assumed to be weak acoustic-structure interactions, the transient structural response is computed first. By applying the FFT, it is transformed into a frequency response. The obtained result is taken as boundary condition for the acoustic part of the vibro-acoustic problem. Consequently, the radiated noise at any point into space can be calculated. The new developed LS-DYNA keyword is based on boundary element method (BEM) in which only the surface of the acoustic domain needs to be discretized. Besides BEM that solves the Helmholtz equation as a linear system, the new card allows, also, to use two other approximative Rayleigh and Kirchhoff methods. Both methods do not require a system of equations to be assembled and solved. Consequently, they are faster than BEM. Rayleigh method assumes that the radiating structure is a plane surface clamped into an infinite rigid plane. In Kirchhoff method, BEM is coupled to FEM used for acoustics in LS-DYNA by prescribing non reflecting boundary condition. In this case, at least one fluid layer needs to be merged to the vibrating structure. Keywords: Acoustic, Vibro-acoustic coupling, FFT, Boundary element method
Matti Loikkanen – Propulsion Engineering Technology & Research, Grama Praveen – Global Research Center, General Electric Company, David Powell – University of California, Berkeley
Ballistic impact on composite panels is studied in this work both experimentally and computationally. The purpose was to develop computational methods to analyze a high speed jet engine fragment impacting on composite targets. 12 inch by 12 inch laminated panels with 8, 16, and 32 plies and the standard +-45, 0 /90 degree stacking sequences were considered. With the nominal ply thickness of 0.0075 inches, the corresponding panel thicknesses were 0.06, 0.125 and 0.25 inches. The panels were mounted on a heavy steel frame and spherical and cylindrical projectiles were shot against composite plates. Several shots with varying impact speeds were fired against each panel thickness. The impact damage was observed, and the initial and exit speeds were measured. The ballistic tests indicated that the amount of energy absorbed during impact by a target is nearly constant showing only a slight increase with increasing initial energy. The amount of energy absorbed per ply increases only slightly for the thicker samples. In addition, the tests showed that the cylindrical projectiles required a larger amount of energy to penetrate the composite panels than did the spherical projectiles LS-DYNA® was used to simulate the tests. The panels were modeled with 8-node solid elements. *MAT_COMPOSITE_DMG_MSC (162) was used to model the orthotropic ply material. This model can be used to model progressive failure of composites with unidirectional and woven fabric fibers. One layer of solid elements was used through the thickness of each ply and several mesh densities were studied. A new Cohesive Contact formulation *CONTACT_AUTOMATIC_SURFACE_TO_SURFACE_TIEBREAK (and ONE WAY TIEBREAK) made available as a “DYCOSS” option was defined between each ply to model delaminations. The cohesive contact – DYCOSS option 9 has been developed during this work. This delamination modeling feature is based on fracture mechanics and requires fracture toughness inputs for the composite material. Option 9 has both the power law and BK-law to account for Mode I and Mode II interaction and allows for a better definition of the constitutive laws. The main advantages of cohesive contact are that it allows the user to toggle very easily between ordinary tie-break based delamination models and cohesive contact models and there is no need for separate cohesive elements in the model. Good overall agreement was found between computations and testing.
Xinran Xiao – General Motors Corporation
Composite tube axial impact is a benchmark problem measuring the predictive capability of composite crash simulation. A previous work revealed that MAT58 in LS-DYNA®, a continuum damage mechanics (CDM) composite material law based on Matzenmiller-Lubliner-Taylor (MLT) model, is inadequate in representing the unloading response of composite tubes that form continuous fronds in axial impact testing. To address the issue, two approaches have been attempted. The first one is to modify the compressive unloading response. The second one is to incorporate plasticity in a CDM framework. Both approaches were tested in an MLT theory based user material model. The modified MLT models improved the stability of the tube crush simulations and the simulation results.
Carla McGregor, Reza Vaziri – The University of British Columbia, Canada, Xinran Xiao – General Motors Corporation
Modeling damage propagation and energy absorption in composite tubular structures under axial compression is a challenging task due to the complex nature of damage growth in composites. In this paper, our model (CODAM for COmposite DAMage), which is incorporated into LS-DYNA® as a user material model, is used to simulate the axial crushing response of braided composite tubes. Recent improvements to the finite element model include the addition of a debris wedge, representation of delamination using a tiebreak contact interface, and more physically based model parameters. It is shown that the damage propagation, fracture morphology and energy absorption predictions correlate well with the experimental results.
Chung-Kyu Park, Seong-Woo Hong, Pradeep Mohan, Richard M. Morgan, Cing-Dao (Steve) Kan – The George Washington University, Kisu Lee, Shinhee Park, Hanil Bae – Hyundai Motor Co. & KIA Motors Corp.
This paper describes the Finite Element (FE) model development of the Progressive Deformable Barrier (PDB). The FE model of the PDB incorporates shear tearing effect and air pressure effect of the honeycomb in the PDB. The developed PDB is used in several vehicle-to-PDB simulations to check its robustness. Five different makes and models of passenger vehicles are selected for the simulation of PDB tests. Finally, the aggressiveness of these vehicles is presented based on the simulation results.
Edwin L. Fasanella, Karen E. Jackson, Sotiris Kellas – NASA Langley Research Center, VA
In June 2007, a 38-ft/s vertical drop test of a 5-ft-diameter, 5-ft-long composite fuselage section that was retrofitted with a novel composite honeycomb Deployable Energy Absorber (DEA) was conducted onto unpacked sand. This test was one of a series of tests to evaluate the multi-terrain capabilities of the DEA and to generate test data for model validation. During the test, the DEA crushed approximately 6-in. and left craters in the sand of depths ranging from 7.5- to 9-in. A finite element model of the fuselage section with DEA was developed for execution in LS-DYNA®, a commercial nonlinear explicit transient dynamic code. Pre-test predictions were generated in which the sand was represented initially as a crushable foam material MAT_CRUSHABLE_FOAM (Mat 63). Following the drop test, a series of hemispherical penetrometer tests were conducted to assist in soil characterization. The penetrometer weighed 20-lb and was instrumented with a tri-axial accelerometer. Drop tests were performed at 16- ft/s and crater depths were measured. The penetrometer drop tests were simulated as a means for developing a more representative soil model based on a soil and foam material definition MAT_SOIL_AND FOAM (Mat 5) in LS- DYNA. The model of the fuselage with DEA was re-executed using the updated soil model and test-analysis correlations are presented.
H. Müllerschön, M. Liebscher – DYNAmore GmbH, Germany, W. Roux, N. Stander – LSTC, U. Reuter – TU Dresden, Germany
For stochastic simulations usually many simulations are performed, therefore much information is available for the simulation engineer. In order to evaluate this information and to assess the results of stochastic investigations soft- ware tools such as LS-OPT and D-SPEX are available. Good and clearly arranged presentation of the results is important, so that the engineer really benefit from the data mining. D-SPEX is intended to provide features that are not currently implemented in the LS-OPT viewer. Therefore, it is a complement to the visualization capabilities of LS-OPT. Its primary focus is on the visualization of meta-models although it also provides features to visualize stochastic results. D-SPEX is also thought of as a testing platform for new features that might evolve in LS-OPT. By opening the command file of LS-OPT, D-SPEX reads all data of the optimization or robustness project. The current version of D-SPEX is fully compatible with LS-OPT version 3.3. For more information on the features of D-SPEX, see .
Mostafa Rassaian, JungChuan Lee, Thomas T. Arakawa – Boeing Phantom Works Structures Technology, Yun Huang, Livermore Software Technology Corporation
Many structures are designed to operate in hot temperature and stringent aero-acoustic fatigue environment, e.g. the engine inlet and the heat shield of aircraft are subject to high temperature and sonic pressure level. It is important to evaluate the dynamic response of the structures exposed to both vibration and acoustic sources of excitations. A new feature of structural analysis with vibro-acoustic loads has been implemented in LS-DYNA®. This feature is based on N-FEARA® finite element analysis tool developed by The Boeing Company. This new capability in LS- DYNA treats the structural response by finite element method coupled with acoustic field based on a known acoustic source behavior by spatial correlation function. The added capabilities enable the users to evaluate the response of structure to both base-excitation, or vibration and acoustic source in the frequency domain. Various acoustic environments and sources of excitations can be considered, including base excitation defining random vibration, in addition to plane wave, progressive wave, reverberant wave, turbulent boundary layer, shock wave, representing various fields for acoustic sources of excitation. Modal acceleration method as well as modal superposition method is used to evaluate the dynamic behavior of structures in the frequency domain. The acceleration power spectral density (PSD) is defined in term of g2/Hz used for vibration analysis. The input spectrum for the acoustic excitations can be either pressure PSD in psi2/Hz or sound pressure level (SPL) in dB. For the latter, sound pressure level is first converted into pressure PSD. The coupling between the structures (represented by the modal shapes) and the acoustic excitations is expressed through the concept of joint acceptance. The results are presented in terms of PSD of the nodal displacements, velocities, accelerations, and element stresses, and the RMS (Root Mean Square) of those variables for a frequency range of interest. Several efficient numerical techniques have been implemented to accelerate the solution phase, including the partition of the panel and subdivision of the range of frequencies. A restart option is provided in case users need to change the input acoustic spectrum or change the range of frequency in the output. Furthermore, this novel feature of LS-DYNA provides users a method to replace an acoustic test environment by a shaker table test as a virtual qualification for testing method. This method is based on a conversion factor between the maximum of root mean square of displacement response due to acoustic pressure load and due to base acceleration load. Several keywords have been introduced in LS-DYNA to facilitate this new feature. Numerical examples are given to demonstrate the new vibro-acoustic analysis capability which will be available in the next release of LS-DYNA.
Marco Perillo, Vito Primavera – EnginSoft SpA, Giorgio Bonello, Marco Cavedoni – Italtractor ITM Spa
ITM Group Engineering Department uses advanced tools as finite element method for static structural analyses of undercarriages, side frames or undercarriage components, as track chain, rollers and tension devices. In order to integrate the recent prototype concepts into this design process combining full system real time dynamic simulations able to represent a typical situation due to operation manoeuvre, experimental test information and 30 years experience of ITM group, a new design procedure is proposed to design and to develop complete undercarriage systems. This paper focuses on the employ of the explicit finite element code LS-DYNA® for predicting reliably the structural behavior of a track chain undercarriage system during usual road obstacles impacts and on the following fatigue life damage analyses of undercarriage (frame components using eta/VPG concept and tools. Part of activity is also developed to investigate how the new simulation procedure could be implemented into ITM Group Engineering Department for increasing design chain efficiency. While the big technical challenge is related to model representation of particular components and to the application of standard operating conditions. Stress and strain field results for full structure and its components are presented and fatigue life of frame principal component is determined and its integrity is evaluated. Finally merits and limits analysis is made in term of quality simulation results, numerical model complexity, design procedure efforts and computational time consuming in comparison to ITM Group Engineering Department experiences.
Bhushan S. Thatte, Gautam S. Chandekar, Ajit D. Kelkar – North Carolina A&T State University, Pramod Chaphalkar – Grand Valley State University
The desired characteristics of materials in modern aircraft applications require high specific modulus and high specific strength for low specific weight. Composite structures, fabricated using carbon and glass fabrics in conjunction with epoxy resin manufactured via the heated vacuum assisted resin transfer molding (H-VARTM) process are now being considered as a low cost alternative to conventional materials without compromising the mechanical properties. It is important to study the response of composite structures under out-of-plane impact loading which may cause considerable damage to the different layers even at low impact energy levels. In the present study, response of composite laminates under low velocity impact loading was investigated using LS-DYNA. The composite laminates were manufactured by the H-VARTM process using basket weave E-Glass fabrics and plain weave AS4 carbon fabrics with the Epon 862 resin system and Epicure-W as a hardening agent. A composite laminate, with 10 layers of carbon and fiberglass fabrics, was modeled using 3D solid elements in a mosaic fashion to represent plain and basket weave patterns. Mechanical properties were calculated by classical micro-mechanical theory and assigned to the elements as orthotropic elastic material properties. The LS-DYNA results were compared with experimental drop test results using the Dynatup Low Velocity Impact Test Machine. The main considerations for comparison were maximum impact load and the energy absorption by the laminates. Progressive damage for fiberglass laminates was reported for six impact energy levels from 128 ft-lbf (incipient damage) to 768 ft-lbf (upper bound) with the increasing increments of 128 ft-lbf. For carbon laminates impact energy levels are half as that of fiberglass.
Tanya Kapoor, William Altenhof – University of Windsor, Andrew Howard – The Hospital for Sick Children
This research focuses on the injury potential of children seated in forward facing child safety seats during side impact crashes in a near side seated position. Side impact dynamic sled tests were conducted by NHTSA at Transportation Research Center Inc. using a Hybrid III 3-year-old child dummy in convertible forward/rearward child safety seat. The seat was equipped with a LATCH and a top tether and the dummy was positioned in forward-facing/near-side configuration. The test was completed using an acceleration pulse with a closing speed of 24.1 km/hr, in the presence of a rigid wall and absence of a vehicle body. A fully deformable finite element model of a child restraint seat, for side impact crash investigations, has been developed which has also been previously validated for frontal and far side impacts. A numerical model utilizing a Hybrid III 3-year-old dummy, employing a similar set-up as the experimental sled test was generated and simulated using LS-DYNA®. The numerical model was validated by comparing the head and the chest accelerations, resultant upper and lower neck forces and moments from the experimental and numerical tests. The simulation results were observed to be in good agreement to the experimental observations. Further numerical simulations were completed employing a rigid ISOFIX system with two different cross-sectional geometries for the anchoring mechanism. It was observed from the simulation results that the use of both rigid ISOFIX geometries was effective in reducing resultant chest accelerations by approximately 40 percent. A reduction of approximately 20 to 30 percent was observed in lateral shear and lateral bending of the dummy’s neck. Of the two rigid ISOFIX geometries considered, the cross-shaped system effectively reduced the lateral head displacement by 27 percent (6.8 cm).
P.A. Du Bois – Consultant, Germany, S. Kolling – Gießen University of Applied Sciences, Germany, M. Feucht – Daimler AG, Germany, A. Haufe – DYNAmore GmbH, Germany
During the past years polymer materials have gained enormous importance in the automotive industry. Especially their application for interior parts to help in passenger safety load cases and their use for bumper fascias in pedes- trian safety load cases have driven the demand for much more realistic finite element simulations. For such applica- tions the material model 187 (i.e. MAT_SAMP-1) in LS-DYNA® has been developed. In the present paper the authors show how the parameters for the rather general model may be adjusted to allow for the simulation of crazing effects during plastic loading. Crazing is usually understood as inelastic deformation that exhibits permanent volumetric deformations. Hence a material model that is intended to be applied for polymer components that show crazing effects during the experimental study, should be capable to produce the correct volu- metric strains during the respective finite element simulation. The paper discusses the real world effect of crazing, the ideas to capture these effect in a numerical model and exemplifies the theoretical ideas with a real world struc- tural component finite element model.
Mitsuhiro Makino – Computational Science and Engineering Solution Center, Fujitsu Limited
In automotive industries, car crash analysis by finite element methods is a very important tool for reducing the development time and cost. In order to get the accurate results, in addition to the improvement of the finite element technology, such as full-integrated shell elements, smaller size of finite element mesh is used, because finer meshes represent the car geometry more accurately, and reduce the noise of contact force. The batch mesh generator, which is enhanced recently, also needs fine mesh. The use of these fine mesh model increases the computational time. In this paper, we examine the performance of the fine mesh model. We developed a 10-million elements car model, which is 10 time larger than the current production car model. The performance of large number of CPU by Massively parallel processing( MPP ) version of LS-DYNA, is measured on Fujitsu PRIMEPOWER.
Eric Stamper, Steven Hale – CAE Associates Inc.
Turbomachinery manufacturers commonly test the centrifugal strength of their rotors in a vertical axis spin test, often called a disk burst test. The design of the containment shell that encloses the disk burst event is critical to ensure the safety of the area surrounding the test. A common method used to design the containment shell for a turbine disk burst test is based on the assumption that the kinetic energy lost by the disk fragments during impact is converted into kinetic energy in the containment shell and energy loss to plastic strain and shear failure in the shell. Containment shells are sized such that the energy required to fail the shell material exceeds the kinetic energy loss during impact. This method is approximate because it assumes fully inelastic impacts and does not account for losses due to friction or heat, nor does it account for stress concentrations in the impact zone or complex disk geometries. ANSYS/LS-DYNA was used to develop an analysis method that could provide more accurate predictions of containment failure limits for a wider range of disk and containment geometries. The ANSYS/LS- DYNA models used a piecewise linear plasticity material law with strain rate dependence, segment based eroding contact , nonlocal failure methods, and a consistent element size. Model results showed good correlation with burst test data  relative to the prediction of containment, shell perforation, and overall deformation.
Mitsuhiro Makino – Fujitsu Limited
The tippe top is a toy consisting of a section of sphere and a short rod. When tippe top, whose rod is at top, is rotating, it automatically inverts and the position of center of mass is raised. This toy is simulated by LS-DYNA.
Brian Walker, Rajab Said – Arup, Philippe Young – Simpleware Ltd
Simpleware have developed a suite of programs that are used to convert imaging data obtained from CT, microCT, MRI or Ultrasound scanning equipment into finite element meshes for use in LS-DYNA. Simpleware provides what is effectively a 3D photocopier: three dimensional replicas can be generated automatically based on scans. In parallel, computer simulations can be used to assess the suitability or performance of objects in operation. Simpleware’s technology has opened up FEA and RP manufacturing to a variety of applications and research fields including: • Industrial reverse engineering • Research in materials and composites • Non-destructive evaluation (NDE) • Biomechanical Research • Implant design and manufacturing • Surgery simulation and planning • Forensics • Biomimicry • Archeology ScanIP is used to import 3D imaging data from MRI, CT, Micro CT and Ultrasound scans. It provides a series of image processing and segmentation tools which allow the user to define areas of interest in the image based on grey scale values. The smoothing algorithms used by ScanIP are volume, topology and geometry preserving. This ensures the accuracy of both the generated surface reconstructions and mesh models is based on image accuracy alone. The segmented areas can then be exported as a 3D stereo lithography file or exported into +ScanFE for meshing. The stereo lithography files can either be used directly for producing rapid prototype parts or imported into CAD software. + ScanCAD allows you to import a CAD model, position it interactively within the 3D imaging data and then generate a Scan IP mask. Scan CAD can be used to obtain patient specific models by positioning CAD models of different implants within a pre-operative scan. Post-operative performance can be simulated using the combined models and multiple scenarios can be tested easily. The paper describes the software and illustrates its use in different fields of application.
Dr.-Ing. Matthias Hörmann – CADFEM GmbH, Germany
Numerical simulations as integral part in the virtual product development process exhibit a huge spectrum. Ranging from simple modal analyses over linear and nonlinear stiffness and strength based problems up to coupled multi- physic analyses, where different physical disciplines interact with each other. Thereby simulation tools in combination with preprocessors must enable users to perform product development tasks faster and therefore more efficient. One essential part is hereby the seamless model file transfer from and to 3D CAD systems. Additionally model and assembly handling in-between the different simulation disciplines in combination with an automatic mesh generation and automatic contact detection is also important to speed up development time. With the Workbench environment, ANSYS took a quantum leap into model analysis and handling different simulation disciplines in one standard user interface in combination with a tight interface from and to almost all common 3D CAD systems. An interface between ANSYS Workbench and LS-DYNA therefore provides the opportunity to use Workbench preprocessing functionalities for LS-DYNA simulations. The German ANSYS and LS-DYNA distributor CADFEM has thus created a unidirectional, interactive graphic interface “Workbench LS-DYNA“ for the transfer of data from ANSYS Workbench to LS-DYNA. This not only enables users to transfer the pure structure in form of nodes and elements, but also sections, materials, contact definitions and boundary conditions, including prescribed motions and force loading. LS-DYNA specific control and database options are included from a template file, which can be customized by the user. Moreover any LS-DYNA keyword command can be defined within the Workbench GUI and will be added into the LS-DYNA input file. Besides using CAD interfaces, automatic mesh generation and contact detection of Workbench, LS-DYNA users will benefit from this interface. With an existing ANSYS Workbench license and LSTC’s free of charge LS-PrePost an additional preprocessor causing cost and training effort may no longer be necessary. Even more important, the interface enables easier data exchange with other analysis departments using already ANSYS Workbench. Moreover the CAD interfaces in ANSYS Workbench also allow a closer link to construction departments.
Zeng-Chan Zhang – Livermore Software Technology Corporation
In this new compressible fluid solver, the conservation element and solution element (CESE) method[1, 2] is used. The CESE method is a novel numerical method for solving conservation laws and it has many nontraditional features, such as: • Space and time conservation ⎯ Flux conservation can be maintained very well both locally and globally in space & time. • Accurate ⎯ It is 2nd order for both flow variables and their spatial derivatives. Thus, it is more accurate than other 2nd order schemes. • Novel & simple shock-capturing strategy ⎯ only a simple weighted averaging technique is used, no Riemann solver and no special limiters are needed to capture shocks. • Both strong shocks and small disturbances can be handled very well simultaneously. Because of these advantages, this CESE solver is a good choice for the following problem simulations: • Compressible flow problems, especially for high speed flows with complex shocks. • Acoustic (noise) problems (near field). This solver is also combined with the LS-DYNA® structure solver to solve the fluid structure interaction (FSI) problems. There, the fluid solver is based on the Eulerian frame while the structure solver is the Lagrangian one. These meshes are independent of each other, and the interfaces will be tracked by the fluid solver automatically. A quasi-constraint method is used in the interface treatment, i.e., the fluid solver get the displacements and velocity of the interfaces from the structure solver and feeds back the fluid pressures (forces) on the interfaces. Currently, both serial & MPP models are available for this compressible fluid & FSI solver (in LS-DYNA® 980 β- version). The fluid mesh can be made up of hexahedra, wedges, tetrahedra, or a mixture of these elements, while the structural mesh can be made up of shells (thin) or solid volume elements. In this talk, a brief review of this new compressible fluid solver will be given first. Then, an introduction of how to use this new solver will be emphasized, including: • Computational domain & mesh determinations, especially for FSI problems • Input deck (keywords cards) setup ⎯ some general control parameters for the method ⎯ initial flow field setup ⎯ boundary condition (BC) choice at each boundary • The use of LS-PrePost® for this new solver’s output In addition, some new features will be introduced, followed by some remarks. The limitations of this solver will be pointed out too. Finally, some features under development will be mentioned.
Ala (Al) Tabiei – Mason, Ohio, Charles Lawrence – NASA Glenn Research Center, Cleveland, OH, Edwin L. Fasanella – NASA Langley Research Center, Hampton, VA
A series of dummy response during a simulated Crew Exploration Vehicle (CEV) module landing is conducted at the Wright-Patterson Air Force Base (WPAFB). These tests consisted of several crew configurations with, and without astronaut suits. Finite element models of the tests are developed and presented herein. The finite element models are validated using the experimental data. Several outputs from the dummy are collected and presented here in. These outputs are compared with the outputs of the finite element model. Occupant crash data such as forces, moments and accelerations are collected from simulations and compared to the presented injury criteria to assess Occupant Survivability and Human Injury. Some of the injury criteria published in the literature are summarized herein for sake of completeness. These injury criteria are used to determine potential injury during such impact event.
Arthur Tang, Jeanne He – Engineering Technology Associates, Inc.
DYNAFORM has been evolved from a Draw Die analysis tool to a Die System analysis tool kits. As the simulation technology and computer resources have been growing rapidly, more demands emerges from different stage of the product and process development sector. Stamping simulation technology is facing more challenges. Based on LS- DYNA ® implicit and explicit solver, DYNAFORM provide simulation tools that support not only the incremental analysis for validation of Draw Die face design, also provides an one-step analysis based cost estimating tool (BSE), Die Face Design tool (DFE) and Die structure analysis, motion transfer and scrap shedding Analysis. DYNAFORM helps the product and process development cycle and makes them more efficient and reliable. Evolving into a process based simulation tool is the future of DYNAFORM. Upgrading the user interface to be flexible for customization and supporting script function are the focus of the next generation DYNAFORM. This paper will also discuss our visions and the future development of DYNAFORM.
Shivakumara H Shetty, Velayudham Ganesan, Suthy C Sivalingam – ESI Group
Visual-Environment (VE) is an open collaborative engineering environment framework or platform called as Open VTOS (Virtual Try-Out Space). VE is an integrated suite of solutions, which has different contexts seamlessly linked for Crash and Safety, Durability, NVH and others. The applications of interest for supporting LS-DYNA based processes are: Visual-Crash DYNA (VCD)-a pre processor for LS-DYNA, Visual-SAFE-an advanced pre-processor for safety features, Visual-Mesh a general purpose mesher, Visual-Viewer (VVI)-a general purpose plotting and simulation application, Visual, Visual-Process Executive-an application for CAE process customization and repetitive tasks automation. These are some of the contexts available in VE but focused to support LS-DYNA. Globalization, new regulations and changes in technologies are influencing the simulation life cycle. These changes are driving the pre and post processing environments for remarkable improvement in productivity, usability and innovative approaches. This paper describes the key features of Visual-Environment 4.0 for LS-DYNA and usefulness of these features in Crash and Safety simulation with productivity examples and process automation.
Velayudham Ganesan, Suthy C Sivalingam, Shivakumara H Shetty – ESI Group, Daniel Dooge – ESI North Americas
Simulation industry has matured enough to carry out the design iterations using just the simulation results without relying on expensive prototype testing. Such simulation approaches have drastically reduced the product development lead time, cost and product failures. Engineers and managers spend most of their time on non engineering activities searching for information, learning simulation tools and workflows, hackneyed repetitive simulations, manual communication and manual simulation data management. The integrated simulation management system such as VisualDSS® (Visual Decision Support System) helps to outwit the current performances of simulation by addressing such non engineering areas of simulation. VisualDSS is a simulation lifecycle management system from ESI’s Visual-Environment suite which aims at providing end-to-end decision support for the simulations such as LS-DYNA® by providing integrated multi disciplinary simulation management features such as simulation data management, compute model management, simulation automation and workflow management, knowledge capture and reuse, simulation project management, results and audit trails processing, smart search and queries, inter enterprises simulation collaboration management and much more. This enables the enterprises to make better decisions, increase the simulation productivity, reduce the project lead time and cost, eliminate the simulation assets loss and improve the simulation reliability. This paper describes the application of VisualDSS in the CAE industry catering multi domain simulation and data management with the objective of supporting decision making process.
Katharina Witowski, Marko Thiele – DYNAmore GmbH, Tushar Goel – Livermore Software Technology Corporation
Most engineering optimization problems require multi-objective optimizations that have no unique optimum because the objectives often conflict. LS-OPT 3.3 offers the capability to simultaneously compute many Pareto optimal solutions using a multi-objective genetic algorithm. The optimization post processing tool D-SPEX provides advanced features to visualize these Pareto optimal solutions and to approximate the Pareto optimal front using the moving least squares method. In addition, the moving least squares method may now be used in D-SPEX as an additional option for building response surfaces as well as for the computation of virtual histories.
Alexander Gromer, Sebastian Stahlschmidt – DYNAmore GmbH, Peter Schuster – Dr. Ing. h.c. F. Porsche AG
The paper describes methodology used to develop the PDB WorldSID model. The test database generated to build and validate the models is described, as well as the design process of a new barrier shape to validate the model in respect to the new FMVSS 214 oblique pole test. Finally, the performance of the current development release is presented.