考虑Sobol缺陷敏感度的抗恶劣环境结构设计方法  

Method for designing parts structure to resist harsh environments considering sobol defects sensitivity

作　　者：尹娇妹 赵昕玥[1] 张树有[1] 

机构地区：[1]浙江大学流体动力与机电系统国家重点实验室机械工程学院,浙江杭州310027

出　　处：《浙江大学学报（工学版）》2015年第8期1487-1494,共8页Journal of Zhejiang University：Engineering Science

基　　金：国家自然科学基金资助项目(51405435);国家"973"重点基础研究发展规划项目计划(2011CB706506)

摘　　要：针对关键零部件结构中一些部位在缺陷对零部件结构的诱变过程中出现过损害现象造成零部件提前报废的问题,提出一种基于sobol缺陷敏感度的结构细节设计分析方法.在关键零部件结构设计的基础上,引入缺陷模型,模拟缺陷对结构的侵袭及迁移过程,以满足结构全局强度为指标,基于sobol缺陷敏感度分析,利用缺陷参数影响零部件结构强度指标程度分离筛选出结构过损害部位,依据较大灵敏度参数选择性的选取优化变量,并利用多目标粒子群算法进行优化,提高零部件结构敏感部位的安全度.算例结果表明所提方法可以有效地改善结构易损部位的应力分布情况,提高零件结构细节抗缺陷的性能.In order to enhance the availability of the vulnerable area,a new method for optimizing mechanical structure to prevent defect was proposed.An abstract defect model was introduced to present the simple invasion and migration process of defect in the structure.Sobol sensitivity technique was used to study the damage degree of parameters in the defect model based on structural analysis.Then the vulnerable areas were labeled by comparing the damage degree.The sensitive areas were taken the object of anti-defect detail structure optimization.The optimization variables were selected tendentiously based on the parameters with bigger damage degree.Multi-objective particle swarm algorithm was employed to find the optimal distribution of the variables in order to improve the safety of the sensitive areas of structure.Results show that the proposed method improves the defect resistance performance of the vulnerable area.

关 键 词：Sobol敏感度分析 缺陷模型 优化变量 粒子群算法 结构细节设计 

分 类 号：TH452[机械工程—机械制造及自动化] TH114