Computing Systems in Engineering最新文献_第7页

A massively parallel implementation of the spectral element method for impact problems in plate structures 板结构冲击问题的谱元方法的大规模并行实现

Computing Systems in Engineering Pub Date : 1994-08-01 DOI: 10.1016/0956-0521(94)90020-5

Albert N. Danial , James F. Doyle

引用次数: 8

Design optimization of large structural systems with substructuring in a parallel computational environment 并行计算环境下具有子结构的大型结构体系设计优化

Computing Systems in Engineering Pub Date : 1994-08-01 DOI: 10.1016/0956-0521(94)90024-8

S.N. Patnaik , A.S. Gendy , D.A. Hopkins

{"title":"Design optimization of large structural systems with substructuring in a parallel computational environment","authors":"S.N. Patnaik , A.S. Gendy , D.A. Hopkins","doi":"10.1016/0956-0521(94)90024-8","DOIUrl":"10.1016/0956-0521(94)90024-8","url":null,"abstract":"<div><p>Design optimization of large structural systems can be attempted through a substructure strategy when difficulties are encountered with convergence of such problems. In this strategy the large structure is divided into several smaller substructures, and a subproblem is defined for each substructure. Solution of the large problem can be obtained iteratively through repeated solutions of the modest subproblems. Substructure strategies, in sequential as well as in parallel computational modes on a CRAY YMP multiprocessor computer, have been incorporated into the code CometBoards, which is an acronym for Comparative Evaluation Test Bed of Optimization and Analysis Routines for Design of Structures. Three issues, intensive computation, convergence of the iterative process, and analytically superior optimum, encountered during the implementation of substructure optimization into CometBoards have been solved. Coupling between subproblems as well as local and global constraints grouping are essential for convergence of the iterative process. The substructure strategy can produce an analytically superior optimum different from what can be obtained by regular optimization. For the problems solved, substructure optimization in a parallel computational mode made effective use of all assigned processors.</p></div>","PeriodicalId":100325,"journal":{"name":"Computing Systems in Engineering","volume":"5 4","pages":"Pages 425-440"},"PeriodicalIF":0.0,"publicationDate":"1994-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1016/0956-0521(94)90024-8","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89756808","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 12

A generalized domain decomposition technique for mixed-iterative finite element formulation 混合迭代有限元公式的广义区域分解技术

Computing Systems in Engineering Pub Date : 1994-08-01 DOI: 10.1016/0956-0521(94)90017-5

Kyung H. Ahn , Dale A. Hopkins

{"title":"A generalized domain decomposition technique for mixed-iterative finite element formulation","authors":"Kyung H. Ahn , Dale A. Hopkins","doi":"10.1016/0956-0521(94)90017-5","DOIUrl":"10.1016/0956-0521(94)90017-5","url":null,"abstract":"<div><p>A concurrent formulation for a large scale finite element problem with direct matrix solver is presented through combined use of domain decomposition and parallel matrix factorization. The algorithm not only reduces high-speed memory requirements but also generates high speed-up in matrix factorization. The present work demonstrates the efficient domain decomposition technique that minimizes the inter-processor communication using the bandwidth optimization technique which has been commonly used for stiffness matrix solution. The direct method using <strong>LDL</strong><sup><em>T</em></sup> skyline factorization has been parallelized and used for each sub-region eliminating the need to form the global system for the entire domain.</p><p>Implementation of the communication algorithm due to domain decomposition into the existing matrix assembly, matrix solution stage along with minimum necessary communication requirements between the processors has been presented in detail. The current domain decomposition scheme provides flexible utilization of processors depending on the user's needs for a problem. The user can assign a different number of processors for domain decomposition and matrix factorization to control the memory requirement and to control the speed-up respectively. The results obtained from IBM RS-6000 clustered workstations will be presented.</p></div>","PeriodicalId":100325,"journal":{"name":"Computing Systems in Engineering","volume":"5 4","pages":"Pages 351-361"},"PeriodicalIF":0.0,"publicationDate":"1994-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1016/0956-0521(94)90017-5","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"76779943","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Advance announcement and call for papers 提前公告并征集论文

Computing Systems in Engineering Pub Date : 1994-08-01 DOI: 10.1016/0956-0521(94)90030-2

引用次数: 0

Postbuckling and large-deflection nonlinear analyses on distributed-memory computers 分布式存储计算机的后屈曲和大挠度非线性分析

Computing Systems in Engineering Pub Date : 1994-08-01 DOI: 10.1016/0956-0521(94)90021-3

Brian C. Watson, Ahmed K. Noor

引用次数: 8

Massively parallel structural design using stochastic optimization and mixed neuralnet/finite element analysis methods 基于随机优化和混合神经网络/有限元分析方法的大规模并行结构设计

Computing Systems in Engineering Pub Date : 1994-08-01 DOI: 10.1016/0956-0521(94)90026-4

Rong C. Shieh

{"title":"Massively parallel structural design using stochastic optimization and mixed neuralnet/finite element analysis methods","authors":"Rong C. Shieh","doi":"10.1016/0956-0521(94)90026-4","DOIUrl":"10.1016/0956-0521(94)90026-4","url":null,"abstract":"<div><p>The title study is performed on the massively parallel processing (MPP) environment of Connection Machine (CM) computers using truss structural sizing design problems as example design problems. In this design optimization procedure, only the displacement solution is replaced by that based on neural net technology (under a given set of cross-sectional size parameters (e.g., areas) in the MPP finite element structural reanalysis). This structural reanalysis procedure, together with a vastly improved and parallelized version of the integral global optimization (IGO) stochastic algorithm, IIGO, forms the present MPP structural design methodology. In addition, a procedure to correct the final optimal design for constraint violation or too-conservatively satisfied constraint condition caused by inaccuracy of the NN (neural network) analysis model is also formulated. Evaluation of the numerical performance of the developed computational algorithm set, capability, and strategy is made, primarily on the Connection Machine CM-2 model computer by performing three neural-network-based truss structural reanalysis/minimum weight design problems.</p></div>","PeriodicalId":100325,"journal":{"name":"Computing Systems in Engineering","volume":"5 4","pages":"Pages 455-467"},"PeriodicalIF":0.0,"publicationDate":"1994-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1016/0956-0521(94)90026-4","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"75872101","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 16

Iterative real/complex eigen-solver and parallel processing for nonlinear panel flutter analysis 非线性板颤振分析的迭代实/复特征求解及并行处理

Computing Systems in Engineering Pub Date : 1994-08-01 DOI: 10.1016/0956-0521(94)90022-1

David Y. Xue , Kotien Wu , Chuh Mei

{"title":"Iterative real/complex eigen-solver and parallel processing for nonlinear panel flutter analysis","authors":"David Y. Xue , Kotien Wu , Chuh Mei","doi":"10.1016/0956-0521(94)90022-1","DOIUrl":"10.1016/0956-0521(94)90022-1","url":null,"abstract":"<div><p>In the frequency domain analysis of the limit cycle motion of a fluttering panel, the operation of a nonlinear eigen-solution is computationally costly. Nonlinear panel flutter analysis includes repeatedly using real and complex eigen-solutions in iterations and in the searching of the stable limit cycle motion. This study presents an efficient iterative real and complex nonlinear eigen-solver to greatly speed up the solution procedure. This new nonlinear eigen-solution adopted a power iteration-scheme and has the following features: (1) it avoids repeatedly using a costly eigen-solver, (2) it is not sensitive to the initial iteration vector, (3) it operates in the real region for a complex solution, and (4) it solves for the single nonlinear mode deflection directly. Those features are particularly suitable for nonlinear panel flutter analysis in which the limit cycle motion is a stable vibration, the eigenvalue and eigenvector are amplitude related and only the dominant eigenvector has a practical meaning.</p><p>The parallel computation is designed to speed up the searching of the stability of the limit cycle motion. The parallel computation was performed by using PVM (Parallel Virtual Machine) on IBM RS/6000 workstations.</p></div>","PeriodicalId":100325,"journal":{"name":"Computing Systems in Engineering","volume":"5 4","pages":"Pages 407-414"},"PeriodicalIF":0.0,"publicationDate":"1994-08-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://sci-hub-pdf.com/10.1016/0956-0521(94)90022-1","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"89240167","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 3

Towards the ultimate iterative substructuring method: Combined numerical and parallel scalability, and multiple load cases 面向终极迭代子结构方法:数值可扩展性与并行可扩展性相结合，多载荷情况

Computing Systems in Engineering Pub Date : 1994-08-01 DOI: 10.1016/0956-0521(94)90016-7

Charbel Farhat, Po-Shu Chen, Paul Stern

引用次数: 12

Parallel three-dimensional mesh generation 平行三维网格生成

Computing Systems in Engineering Pub Date : 1994-08-01 DOI: 10.1016/0956-0521(94)90014-0

Hugues L. de Cougny, Mark S. Shephard, Can Ozturan

引用次数: 40

Optimal design of laminated composite structures using coarse-grain parallel genetic algorithms 基于粗粒并行遗传算法的层合复合材料结构优化设计

Computing Systems in Engineering Pub Date : 1994-08-01 DOI: 10.1016/0956-0521(94)90023-X

W.F. Punch III , R.C. Averill , E.D. Goodman , S.-C. Lin , Y. Ding , Y.C. Yip

引用次数: 31