首页 | 本学科首页   官方微博 | 高级检索  
     

基于OpenSees的大型结构分析GPU高性能计算方法
引用本文:解琳琳,;韩博,;许镇,;陆新征,;任爱珠.基于OpenSees的大型结构分析GPU高性能计算方法[J].土木建筑工程信息技术,2014,6(5):22-25.
作者姓名:解琳琳  ;韩博  ;许镇  ;陆新征  ;任爱珠
作者单位:1.清华大学土木工程系, 北京 1000842.北京市建筑设计研究有限公司, 北京 100045
基金项目:国家科技支撑计划课题(2013BAJ08B02); 国家自然科学基金(91315301,51378299); 北京市自然科学基金(8142024)
摘    要:近年来我国大型结构抗震弹塑性分析得到了广泛应用, 其中OpenSees作为一款开源有限元程序已逐渐成为最具国际影响力的计算平台之一。但是由于该软件的矩阵求解器效率较低, 难以满足大型结构的计算需求。因此本文以GPU高性能求解为技术手段, 在OpenSees中集成了两个基于迭代法的并行线性求解器, 并通过两个大规模模型验证了基于GPU进行大型结构弹塑性分析的可靠性和高效性(加速比9~15), 为进一步开展基于OpenSees的大型结构地震灾变行为研究提供了高效的计算方法。

关 键 词:大型结构  弹塑性分析  OpenSees  GPU  并行迭代线性求解器

GPU Powered High-performance Computing Method for the Analysis of Large-Scale Structures based on OpenSees
Affiliation:Xie Linlin,Han Bo,Xu Zhen,Lu Xinzheng,Ren Aizhu(1. Department of Civil Engineering Tsinghua University, Beijing 100084, China; 2. Beijing institute of architecture design, Beijing 100045, China)
Abstract:Numerical simulation using finite element(FE) software packages has been widely adopted to investigate the seismic performance of large-scale structures,recently. Among these,Open System for Earthquake Engineering Simulation(OpenSees),as an open-source FE software program,has increasingly become one of the most influential packages. However,the computational efficiency of the matrix solvers currently cannot satisfy the demand for numerical simulation of large-scale structures. Consequently,two new,parallel,iterative linear solvers are proposed and implemented in Open Sees based on two graphic processing unit-based( GPU-based) libraries. The time history analysis of two typical large-scale structures are performed using the proposed solvers. The speed-up ratio of the proposed solvers are up to 9 to 15(achieved with high accuracy in results). The research outcome can provide an effective computing technology for further numerical analysis of the seismic behavior of large-scale structures.
Keywords:Large-scale Structures  Numerical Analysis  OpenSees  GPU  Parallel Iterative Linear Solver
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号