点击次数：

论文类型： 会议论文

发表时间： 2013-01-01

收录刊物： CPCI-S、Scopus

卷号： 1532

页面范围： 1073-1078

关键字： Neural Network; Body Design; Joint; Finite Element

摘要： In order to estimate the mechanical properties of the overall structure of the body accurately and quickly in conceptual design phase of the body, the beam and shell mixing elements was used to build simplified finite element model of the body. Through the BP neural network algorithm, the parameters of the mechanical property of joints element which had more affection on calculation accuracy were calculated and the joint finite element model based on the parameters was also constructed. The case shown that the method can improve the accuracy of the vehicle simulation results, while not too many design details were needed, which was fit to the demand in the vehicle body conceptual design phase.