摘要
有限元分析作为一种重要的计算机辅助工程技术,在汽车零部件开发中起着极为重要的作用,但是在分析复杂问题时耗时长,影响开发周期。为此,本文提出了一种以有限元仿真提供样本数据、以图网络技术建立有限元输入与输出间的映射关系的神经网络方法,将该图网络方法应用于座椅骨架总成的应力场预测。该预测方法中采用图节点和图边模拟了有限元模型中节点间的连接关系,由此表达有限元模型中单元之间的拓扑关系,并将预测结果与有限元仿真结果进行了对比。结果表明该方法能准确预测座椅骨架总成的最大应力及其出现的位置,对应力分布一致性均有较好的预测能力,且计算速度比相应有限元求解度快3个数量级以上。
Finite Element Analysis(FEA),as an important Computer-aided Engineering(CAE)technolo⁃gy,plays a significant role in the area of automotive part development.However,it costs too much time when solv⁃ing complicated problems,which affects the development cycle.In this paper,a neural network method is pro⁃posed,in which sample data is provided by finite element simulation and the mapping relationship between finite el⁃ement input and output is established by graph network technology.The graph network method is used to predict the stress field of the seat frame assembly.The prediction method simulates the connection relationship between nodes in the finite element model using graph nodes and graph edges,which can effectively express the topological rela⁃tionship between elements in the finite element model.The prediction results are compared with the results of the fi⁃nite element simulation.The results show that the method can precisely predict the maximum stress and its corre⁃sponding location of the seat frame assembly,with strong predictive capabilities for stress distribution consistency.Additionally,the model has a significant computational advantage,with a calculation speed three orders of magni⁃tude faster than that of the corresponding finite element solver.
作者
高泽
楚遵康
石稼晟
林滏
饶卫雄
余海燕
Gao Ze;Chu Zunkang;Shi Jiasheng;Lin Fu;Rao Weixiong;Yu Haiyan(School of Automotive Studies,Tongji University,Shanghai 201804;School of Software,Tongji University,Shanghai 201804)
出处
《汽车工程》
EI
CSCD
北大核心
2024年第1期170-178,共9页
Automotive Engineering
基金
国家重点研发计划项目(2022YFE0208000)资助。
关键词
有限元分析
深度学习
图网络
汽车座椅
finite element analysis
deep learning
graph network
automotive seat
