具有强非线性状态方程的机械零部件可靠性灵敏度分析方法

Advanced Method to Estimate the Reliability-Based Sensitivity of Mechanical Components With Strongly Nonlinear Performance Function

基于可靠性灵敏度设计的随机摄动技术,结合可靠性分析的矩方法、矩阵微分理论和Kro-necker代数的相关理论,讨论了实际中存在着高度非线性极限状态方程结构的可靠性灵敏度问题.在已知随机变量前4阶矩的前提下,对基于摄动法的可靠性灵敏度计算方法进行了修正,提出了具有高度非线性结构的可靠性灵敏度计算方法.并结合实例证明了采用此方法大大提高了可靠性灵敏度的计算精度,并为工程实际提供了更加可信的理论依据.

Based on the random perturbation technique of reliability sensitivity design, some realistic reliability-based sensitivity issues were discussed, with some of them having a structure of high nonlinear performance function. Combining the relating theories of moment method of reliability analysis, matrix differential and Kronecker algebra, the reliability-based sensitivity method based on perturbation method was modified, given that the first four moments of random variables. Meanwhile, a reliability-based sensitivity computing methodology also was put forward. Some examples were adopted to prove that using this methodology could highly improve the accuracy of reliability-based sensitivity computation and this methodology also offers some more dependable theoretic basis for engineering.