首页 | 本学科首页   官方微博 | 高级检索  
     

汽车碰撞过程的计算负载均衡并行有限元技术
引用本文:张迎军,吴沈荣,施欲亮,涂金刚,李文平.汽车碰撞过程的计算负载均衡并行有限元技术[J].计算机辅助工程,2017,26(3):13-18.
作者姓名:张迎军  吴沈荣  施欲亮  涂金刚  李文平
作者单位:奇瑞汽车股份有限公司 汽车工程技术研发总院,奇瑞汽车股份有限公司 汽车工程技术研发总院,奇瑞汽车股份有限公司 汽车工程技术研发总院,奇瑞汽车股份有限公司 汽车工程技术研发总院,奇瑞汽车股份有限公司 汽车工程技术研发总院
摘    要:为了在有限的计算资源下挖掘更高的汽车碰撞仿真计算效率,基于当前普遍采用的并行处理区域分解算法,结合一个典型碰撞案例分析可知,汽车碰撞模型计算过程中的主要时间开销与结构大变形相关.提出基于坐标递归对分(Recursive Coordinate Bisection,RCB)的计算负载均衡区域分解算法,并将此法应用于某整车碰撞仿真模型的3类典型工况.计算结果表明此法能将并行计算效率提高10%~24%.

关 键 词:汽车碰撞仿真    并行计算    区域分解    计算负载均衡    坐标递归对分    计算效率
收稿时间:2017/2/12 0:00:00
修稿时间:2017/3/12 0:00:00

Computational load balancing for parallel explicit finite element technology of vehicle impact
ZHANG Yingjun,WU Shenrong,SHI Yuliang,TU Jingang and LI Wenping.Computational load balancing for parallel explicit finite element technology of vehicle impact[J].Computer Aided Engineering,2017,26(3):13-18.
Authors:ZHANG Yingjun  WU Shenrong  SHI Yuliang  TU Jingang and LI Wenping
Affiliation:Advanced and Core Engineering Institute, Chery Automobile Co., Ltd.,Advanced and Core Engineering Institute, Chery Automobile Co., Ltd.,Advanced and Core Engineering Institute, Chery Automobile Co., Ltd.,Advanced and Core Engineering Institute, Chery Automobile Co., Ltd. and Advanced and Core Engineering Institute, Chery Automobile Co., Ltd.
Abstract:To improve the efficiency of simulation calculation of vehicle impact from the limited computational resources, based on the commonly used parallel processing domain decomposition algorithm, a typical impact case is analyzed, and the results show that the main time consumption of automobile impact simulaiton is correlated with the large deformation of structure. A domain decomposition algorithm for computational load is proposed on the basis of Recursive Coordinate Bisection(RCB). The method is applied in three typical cases of a vehicle impact simulation model, and the computational results show that the method is able to improve parallel computing efficiency by 10% to 24%.
Keywords:vehicle impact simulation  parallel computing  domain decomposition  computational load balance  recursive coordinate bisection  computational efficiency
点击此处可从《计算机辅助工程》浏览原始摘要信息
点击此处可从《计算机辅助工程》下载全文
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