Parallel computation methods for large-scale nonlinear CSM

Parallel computation methods for large-scale nonlinear CSM

This work describes scalable coupling of two stand alone computer codes for multiscale impact analysis of composites. An asymptotic expansion homogenization (AEH) based microstructural code available for modeling microstructural aspects of modern armor materials is coupled with PARADYN, a parallel explicit Lagrangian finite element code. The first code enables modeling of material microstructures in simple loading situations in stand-alone form. The coupling of this code to PARADYN, which is a parallel version of the Lawrence Livermore National Laboratory (LLNL)'s serial DYNA3D, enables a micro-macro type multiscale analysis of large elastic-plastic deformation response of structures under generalized three dimensional impact conditions. Three sets of results are presented to demonstrate: 1) the verification of the AEH-PARADYN model coupling to PARADYN, 2) validation of the AEH, and 3) the scalability of the coupled model.

Salinas, a massively parallel implicit finite element code developed at Sandia National Laboratories under the Accelerated Strategic Computing Initiative (ASCI), has demonstrated phenomenal almost linear speedup factors up to tens of thousands of processors. However, its use so far has been limited to Department of Energy (DOE) laboratories and government entities. The present work investigates the ease of use and applicability of Salinas from an aerospace designer’s perspective, as opposed to a researcher’s viewpoint. Salinas is exercised on a number of relatively simple static and modal, linear and large deformation analysis problems, and the results are compared to NE/NASTRAN. Overall, the results obtained with the two codes match very closely, once the user gets familiar with Salinas to be able to understand its particularities. Subsequently, two shell models representative of typical aerospace structures are analyzed, a circular bulkhead with a pressure load and a design model of a STOVL fighter aircraft. Again, once the NASTRAN models are modified such that they translate into Salinas without error, very closely matching results are obtained. While Salinas does not currently support all of the features of NASTRAN which are used by a designer, it does not seem to be a far stretch to assume that all this functionality can relatively easily be added to Salinas, turning it into a viable design tool able to significantly reduce analysis time when run on a massively parallel architecture.

Dynamic delamination modelling using interface elements

The simulation of damage process of high arch dam subjected to strong earthquake shocks is significant to the evaluation of its performance and seismic safety, considering the catastrophic effect of dam failure. However, such numerical simulation requires rigorous computational capacity. Conventional serial computing falls short of that and parallel computing is a fairly promising solution to this problem. The parallel finite element code PDPAD was developed for the damage prediction of arch dams utilizing the damage model with inheterogeneity of concrete considered. Developed with programming language Fortran, the code uses a master/slave mode for programming, domain decomposition method for allocation of tasks, MPI (Message Passing Interface) for communication and solvers from AZTEC library for solution of large-scale equations. Speedup test showed that the performance of PDPAD was quite satisfactory. The code was employed to study the damage process of a being-built arch dam on a 4-node PC Cluster, with more than one million degrees of freedom considered. The obtained damage mode was quite similar to that of shaking table test, indicating that the proposed procedure and parallel code PDPAD has a good potential in simulating seismic damage mode of arch dams. With the rapidly growing need for massive computation emerged from engineering problems, parallel computing will find more and more applications in pertinent areas.

A micro-mechanical model is developed for laminated composite materials and implemented in the explicit finite element method. The objective of this study is to get an accurate and simple micro-model, which can be used in the displacement-based explicit finite element packages. The micro-mechanical model implemented in the explicit finite element code can be used for simulating the behavior of composite structures under various loads such as impact and crash. The stress-strain relation for the micro-model is derived for shell element. As a demonstration case of the developed micro-model a finite element model of Graphite/Epoxy tube structure is developed and simulated under axial crash.

Laminated glass is widely used for architectural glazing and as windshield material in automotive industry. The architectural application of laminated glass is subjected to low-velocity impact by flying debris in hurricane, tornado, and blast events. The automotive application is subjected to low-velocity impact in traffic accidents. Circular samples of laminated glass are subjected to low-velocity impact at 1, 2, 3, and 4 m/s impact velocity approximately by means of Instron Dynatup testing machine. The impact tup force is recorded by data acquisition equipment of the machine. The low-velocity impact of laminated glass is modelled in explicit finite elements, using the commercial software ABAQUS. Three finite element models are examined: brick element model with 4 plies of elements per a glass layer and one per the interlayer, continuum shell element model with a brick element ply for the interlayer, and conventional shell element model excluding the interlayer shear. The glass layers are modelled by brittle damage material model, which is based on fracture toughness Charpy tests. The interlayer of the polymeric (PVB) material is modelled as elastic material with instantaneous Young’s modulus found in a creep test. The careful modelling of laminated glass low-velocity impact in explicit finite element code provides corresponding to the experiment results. The conventional shell element model underestimates the energy dissipation. The continuum shell element model could be very efficient if it is stable enough. The necessity of efficient laminated glass shell element is discussed. The finite element analysis gives an insight of the progressive failure of the laminated glass, explaining the place and the order of radial and circular crack appearance. The investigation of the laminated glass response to the low-velocity impact gives the opportunity to optimize the laminated glass layers for higher resistance.

The application of parallel computers to reservoir simulation is studied with the aid of two recently introduced Multiple Instruction, Multiple Data (MIMD) computers. The problems posed by reservoir simulators are shown to be particularly well suited for such parallel computers* Fundamental parallel programming techniques using Fortran are presented for the Hetrogeneous Element Processor (HEP) produced by Denelcor and the iPSC Hypercube produced by Intel. These techniques are then applied to two of the most time consuming tasks in reservoir simulation: forming the matrix coefficients and producing a sparse matrix solution. The problem of parallel formation of sparse matrix coefficients is addressed for the single and multiphase cases using both black oil and compositional fluid models. Parallel sparse matrix solution is addressed by considering D4 ordered Gaussian elimination, and the multigrid, conjugate gradient, and Successive Over-Relaxation (SOR) methods. Three parallel algorithms based on SOR are developed to illustrate the possible diversity of parallel algorithms. The red-black and multicolored SOR methods which are often used on vector machines are shown to be easily adaptable to parallel MIMD machine. Matrix partitioning and a new method of iteration pipelining are also presented as parallel SOR methods. The various SOR algorithms, in parallel form, are mapped onto the HEP to illustrate the speedups currently possible on an MIMD machine. Projections are made on the future importance of these types of machines to reservoir simulation.

Virtual earthquake engineering laboratory with physics-based degrading materials on parallel computers

Computing the stresses and deformations of the human eye components due to a high explosive detonation using fluid–structure interaction model

Abstract: A fracture energy approach for modelling mixed‐mode delamination of composite materials and other bonded structures is introduced. The model is incorporated within an explicit finite element (FE) code and ties layered shell elements together via a stiffness condition, a failure criterion and post‐failure damage law. The procedure for predictive modelling of delamination using the approach is described and the set of required input parameters is presented. A benchmark test comparing experimental results for a continuous filament random E‐glass/polyester composite and explicit FE simulations for standard fracture toughness tests for a range of mode mixities is included.

The difficulty with an OOP situation compared to an in-position situation is that the inflation of the folded airbag is much more important. It has to be realistically computed, since it affects the impact of the dummy. Attaining a realistic simulation means starting with a correct geometry of the folded airbag and simulating the inflation with correct gas dynamics. Several commercial software packages exist that can simulate the inflation process of an airbag, e.g., the explicit Finite Element (FE) code LS-DYNA [5].

A nonlinear higher order shear deformation shell element for dynamic explicit analysis:: Part I. Formulation and finite element equations

Crashworthiness design draws considerable attention and interest in energy absorption systems. An optimal design of circular stepped tubes based on free inversion is presented in the current work to improve and enhance the crashworthiness characteristics under axial impact loading. Circular tubes have been known to be efficient energy absorbers per unit weight, and may be considered as the main part of energy absorbing devices even for frequently impact operation like landing gear or for sudden accidents like other vehicles. The precise choice of the circular tubes' material, shape, and mechanism may save a person life or save a valuable equipment from damage. Adaptive Multiple-Objective Optimization (AMO) can obtain more light and efficient section. An optimization of variable parameters, which represent the shape of a stepped tube under axial impact, is performed using the platform of ANSYS-DESIGN-EXPLORATION linked with LSDYNA code. The analysis leads to 16% reducing in mass with corresponding 31% improving in energy absorption.

Sheet hydroforming of woven FRT composites: non-orthogonal constitutive equation considering shear stiffness and undulation of woven structure