自适应有限元常用标记策略及其在PHG中的并行实现

讨论自适应有限元计算中常用的标记策略的并行实现问题,介绍并行自适应有限元软件平台PHG中实现这些策略的统一函数接口.特别地,针对一类在分布式存储并行计算机上不易实现的策略,如GERS策略和MNS策略,介绍我们所设计的并行算法.     

Three dimensional finite element modeling of thermomechanical frictional contact between finite deformation bodies using R-minimum strategy

An algorithm for analyzing the transient thermal coupling with the frictional contact between the multiple elastic–plastic bodies in finite deformation is presented using the R-minimum strategy. An arbitrarily shaped contact element strategy, named as node-to-point contact element strategy, is proposed to handle the thermomechanical frictional contact between finite deformation bodies. Assuming the material properties to be temperature dependent, the constitutive equations for both the thermomechanical frictional contact and the thermal-elastic–plastic materials are deduced respectively and applied in our finite element code. Finally, two examples are presented to show the efficiency and usefulness of this algorithm.     

Explicit nonlinear dynamic finite element analysis on homogeneous/heterogeneous parallel computing environment

This paper presents parallel computational strategies to implement explicit nonlinear finite element analysis code onto distributed memory parallel computers for solving large-scale problems in structural dynamics. Implementation details on both homogeneous and heterogeneous parallel processing environments are considered in detail in this paper. Implementation of an explicit nonlinear finite element dynamic analysis code on homogeneous systems is discussed first and this is later moved onto heterogeneous systems. Domain decomposition with explicit message passing is preferred for parallel implementation. The message passing implementation in the parallel algorithm is based on MPI (Message Passing Interface) libraries. Implementation aspects of overlapped, non-overlapped domain decomposition techniques, Dynamic Task Allocation (DTA) and clustering techniques for DTA and their relative merits are presented. The interprocessor communications are optimised by overlapping with computations to improve the performance of the domain decomposition based explicit dynamic analysis finite element code.The issues related to implementation of finite element code for nonlinear dynamic analysis on heterogeneous parallel computing environment are later presented. A new dynamic load-balancing algorithm is developed for this purpose and it is integrated with the domain decomposition based parallel explicit finite element code to test our algorithms on a coarse grain heterogeneous cluster of workstations. Numerical experiments have been carried out on PARAM-10000, an Indian parallel computer and also on cluster of Unix workstations.     

A parallel finite element method and its prototype implementation on a hypercube

This paper describes a parallel implementation of the finite element method on a multiprocessor computer. The proposed strategy does not require the formation of global system equations. An element or substructure is mapped onto each processor of the multiple-instruction, multiple-data multiprocessing system. Throughout the program, each processor stores only the information relevant to its element (substructure) and generates the local stiffness matrix. A parallel element (substructure) oriented conjugate gradient procedure is employed to compute the displacements. Each processor then determines the strains and stresses for its associated element (substructure). A prototype implementation of this parallel finite element program strategy on a hypercube computer is discussed. Examples for both linear and nonlinear analyses are presented.     

A finite element method for a class of contact-impact problems

We present a finite element method for a class of contact-impact problems. Theoretical background and numerical implementation features are discussed. In particular, we consider the basic ideas of contact-impact, the assumptions which define the class of problems we deal with, spatial and temporal discretizations of the bodies involved, special problems concerning the contact of bodies of different dimensions, discrete impact and release conditions, and solution of the nonlinear algebraic problem. Several sample problems are presented which demonstrate the accuracy and versatility of the algorithm.     

A high performance crashworthiness simulation system based on GPU

Crashworthiness simulation system is one of the key computer-aided engineering (CAE) tools for the automobile industry and implies two potential conflicting requirements: accuracy and efficiency. A parallel crashworthiness simulation system based on graphics processing unit (GPU) architecture and the explicit finite element (FE) method is developed in this work. Implementation details with compute unified device architecture (CUDA) are considered. The entire parallel simulation system involves a parallel hierarchy-territory contact-searching algorithm (HITA) and a parallel penalty contact force calculation algorithm. Three basic GPU-based parallel strategies are suggested to meet the natural parallelism of the explicit FE algorithm. Two free GPU-based numerical calculation libraries, cuBLAS and Thrust, are introduced to decrease the difficulty of programming. Furthermore, a mixed array and a thread map to element strategy are proposed to improve the performance of the test pairs searching. The outer loop of the nested loop through the mixed array is unrolled to realize parallel searching. An efficient storage strategy based on data sorting is presented to realize data transfer between different hierarchies with coalesced access during the contact pairs searching. A thread map to element pattern is implemented to calculate the penetrations and the penetration forces; a double float atomic operation is used to scatter contact forces. The simulation results of the three different models based on the Intel Core i7-930 and the NVIDIA GeForce GTX 580 demonstrate the precision and efficiency of this developed parallel crashworthiness simulation system.     

基于龙格库塔法的弹塑性有限元并行计算

基于MPI集群环境对弹塑性区域分解有限元并行计算进行研究。提出了基于三阶和四阶的龙格库塔（Runge-Kutta）方法对应力-应变关系进行积分的算法。积分过程中自动调整子步大小来控制积分过程中的误差。研制了采用最小残余平滑法的子结构预处理共轭梯度并行求解算法。算法在基于工作站机群的并行环境下实现。计算结果表明：该算法具有良好的并行加速比和效率,是一种有效的并行求解算法。     

Efficient sensitivity analysis and optimization of shell structures by the ABAQUS code

In this paper, a new efficient sensitivity analysis procedure is presented for the optimization of shell structures without access to the finite element source code. It is devised as a general interface tool to extend existing finite element systems from pure structural analysis to design capability. The implementation is performed based on the ABAQUS code. Kirchhoff flat shell elements are taken into account in the study with the element thickness as design variables. To ensure the performance and the validity of the proposed procedure, satisfactory sensitivity and optimization results are illustrated for numerical examples.     

An algorithm for isochrome raster graphic display

Raster graphic display is used increasingly with microcomputers making use of readily available television monitors. An algorithm for obtaining the variation of a function on a screen by showing isochrome segments is presented. Small isoparametric bilinear elements are used as an adjustable intermediate size between the finite element mesh itself and the pixel. This approach permits a simple computer code that is adjustable to different computer hardware. A flow chart of the algorithm is presented and errors associated with the display are discussed. The results of two analyses are given. These examples were chosen for their differences both in the type of problem and in the numerical methods of solution used. First, the stresses in a gear tooth submitted to a concentrated load are presented. In this case, the algorithm is part of a post-processor following a finite element displacement analysis. A second example illustrates stream lines in a channel with a step. For the latter, the display shows the output data from a Gauss-Siedel resolution of a Laplace equation using a finite difference mesh.     

Non-intrusive Coupling: Recent Advances and Scalable Nonlinear Domain Decomposition

This paper provides a detailed review of the global/local non-intrusive coupling algorithm. Such method allows to alter a global finite element model, without actually modifying its corresponding numerical operator. We also look into improvements of the initial algorithm (Quasi-Newton and dynamic relaxation), and provide comparisons based on several relevant test cases. Innovative examples and advanced applications of the non-intrusive coupling algorithm are provided, granting a handy framework for both researchers and engineers willing to make use of such process. Finally, a novel nonlinear domain decomposition method is derived from the global/local non-intrusive coupling strategy, without the need to use a parallel code or software. Such method being intended to large scale analysis, we show its scalability. Jointly, an efficient high level Message Passing Interface coupling framework is also proposed, granting an universal and flexible way for easy software coupling. A sample code is also given.