结构时变失效概率函数估计的加权重要抽样方法  

WEIGHTING IMPORTANCE SAMPLING METHOD FOR STRUCTURAL TIME-DEPENDENT FAILURE PROBABILITY FUNCTION ESTIMATION

作　　者：钱宇耕 袁修开 陈敬强 QIAN YuGeng;YUAN XiuKai;CHEN JingQiang(School of Aerospace Engineering,Xiamen University,Xiamen 361005,China)

机构地区：[1]厦门大学航空航天学院,厦门361005

出　　处：《机械强度》2023年第6期1387-1393,共7页Journal of Mechanical Strength

基　　金：航空科学基金(20170968002)项目资助。

摘　　要：参数时变失效概率函数(Time-Dependent Failure Probability Function,TDFPF)是结构时变可靠性优化设计的关键问题之一,其估计的精度与效率决定了结构时变可靠性优化的效果。针对时变失效概率函数计算成本高的问题,基于结构复合极限状态思想提出了基于加权重要抽样的结构时变失效概率函数估计策略。对于加权重要抽样进行扩展,仅需一次可靠性分析,即可得到时变失效概率函数的估计,避免了多次重复计算。为了验证所提方法的精度与效率,给出数值算例及工业机器人减速器结构时变算例。结果表明,对于时变失效概率函数估计问题,所提方法适用性好,能够以较高效率获得满意的结果。Time-dependent failure probability function(TDFPF)of parameters is one of the key problems of structural timedependent reliability-based optimization.The accuracy and efficiency of TDFPF estimation determine the effect of time-dependent reliability-based optimization of structures.In this work,a weighted importance sampling(WIS)approach was proposed for the TDFPF estimation based on the concept of the composite limit states.The proposed method extends importance sampling to obtain the estimation of the TDFPF with only one reliability analysis,avoiding repeated reliability analyses.To verify the accuracy and efficiency of the proposed method,numerical examples and industrial robot reducer examples are given and solved by the proposed method.The results indicated that the proposed method has wide applicability,high efficiency in obtaining the satisfactory estimation of the TDFPF.

关 键 词：时变可靠性 时变失效概率函数 加权策略 重要抽样法 随机模拟 

分 类 号：TB114.3[理学—概率论与数理统计]