薄壁结构抗弯性能多目标优化设计方法研究  

Research on multi-objective optimization design for flexural performance of thin-walled structure

作　　者：徐小东[1] 谷先广[2] 

机构地区：[1]安徽交通职业技术学院汽车与机械工程系,安徽合肥230051 [2]合肥工业大学机械与汽车工程学院,安徽合肥230009

出　　处：《合肥工业大学学报（自然科学版）》2015年第4期446-449,462,共5页Journal of Hefei University of Technology：Natural Science

基　　金：合肥工业大学博士学位人员专项基金资助项目(2013HGBZ0164)

摘　　要：薄壁吸能结构的抗弯性能评估是多方面的,并且各个质量特性之间可能会发生相互冲突。因此,薄壁结构弯曲吸能特性的优化设计非常困难。为了解决上述问题,文章提出了一种结合试验设计技术、径向基函数近似建模技术、序列采样策略和多目标粒子群算法的序列近似多目标优化设计方法,进行了薄壁结构的弯曲吸能特性优化设计。优化结果表明,该方法提高了薄壁结构的弯曲吸能特性,比传统的优化方法具有更高的优化设计效率;并且在相同的求解效率下,序列采样策略比常规采样策略具有更高的精度,能大幅提高产品的设计质量。The flexural performance of thin-walled energy absorption structures is characterized by a number of quality and/or performance indices,some of which could conflict with each other.Therefore,it would be very difficult to make the optimization design of the bending energy absorbing ability of thin-walled structures.In order to solve the problem above-mentioned,a sequential approximation multi-objective optimization algorithm is presented which integrates experimental design technology,radial basis function approximation modeling technology,sequential sampling strategy and multi-objective particle swarm optimization algorithm.The bending energy absorbing ability of the thin-walled structures is optimized by the presented method,and the results show that the method can significantly improve the bending energy absorbing ability of thin-walled structures.The method has higher optimization efficiency than traditional optimization methods.And the sequential sampling method has higher accuracy than the conventional sampling strategy under the same solution efficiency,thus greatly improving the design quality of the product.

关 键 词：径向基函数 薄壁结构 多目标优化 序列近似 

分 类 号：TG386[金属学及工艺—金属压力加工]