基于改进多层级嵌套Copula方法的机电产品可靠性评估  

Reliability evaluation for electromechanical products based on improved multi-level nested Copula method

作　　者：苏春[1] 汪千程 Su Chun;Wang Qiancheng(School of Mechanical Engineering,Southeast University,Nanjing 211189,China)

机构地区：[1]东南大学机械工程学院,南京211189

出　　处：《东南大学学报（自然科学版）》2022年第5期981-989,共9页Journal of Southeast University：Natural Science Edition

基　　金：国家自然科学基金资助项目(71671035);机械设备健康维护湖南省重点实验室开放基金资助项目(201901);江苏风力发电工程技术中心开放基金资助项目(ZK19-03-03)。

摘　　要：考虑系统内部和外部相关性变量的共同作用,提出一种基于改进多层级嵌套Copula的可靠性评估方法.将机电产品分解为系统级、子系统级和零部件级3个层级,根据故障率模型拟合各零部件的故障概率,并作为其边缘分布函数;通过计算边缘分布之间的秩相关系数,确定零部件之间的故障相关性;利用改进多层级嵌套Copula完成边缘分布的逐层更新和参数估计,建立零部件级的故障相依结构,通过递归得到子系统可靠性模型;重复上述步骤,建立子系统级的故障相依结构,得到系统级可靠性模型,实现多变量、多层级产品可靠性评估.以某型号浮式海上风力机为例,验证方法的有效性.结果表明,改进多层级嵌套Copula方法能准确定义多元变量、不同层级之间的非线性相关关系,消除传统嵌套Copula方法在分层处理时存在的误差,提高机电产品可靠性评估的准确性.On the basis of the improved multi-level nested Copula, a reliability evaluation method was presented, considering the joint action of internal and external correlation variables. The electromechanical products were divided into three levels, system level, subsystem level and component level. The failure probability for each component was fitted by the failure rate model, and taken as its edge distribution function. By calculating the rank correlation coefficient between the edge distributions, the fault correlation between the components was determined. Moreover, the improved multi-level nested Copula was used to update the edge distribution layer by layer and estimate the parameters, the failure dependence structure at the component level was established, and the reliability model of the subsystem was obtained recursively. Repeating the above steps, the failure dependence structure at the subsystem level was established, and the reliability model was obtained for multi-variable and multi-level products finally. A floating offshore wind turbine was used as an example to illustrate the effectiveness of the proposed method. The results show that the improved multi-level nested Copula method can accurately define the nonlinear correlation between multivariate variables and different levels, which is conducive to eliminate the error of traditional nested Copula method in layered processing, thus improving the accuracy of reliability evaluation.

关 键 词：多层级嵌套Copula 故障相关性 层次相关性 可靠性评估 

分 类 号：TH17[机械工程—机械制造及自动化]