吉林大学学报(理学版) ›› 2022, Vol. 60 ›› Issue (2): 284-294.

• • 上一篇    下一篇

Burr Ⅻ部件相依屏蔽数据系统的可靠性分析

李玥, 冶继民   

  1. 西安电子科技大学 数学与统计学院, 西安 710126
  • 收稿日期:2021-06-28 出版日期:2022-03-26 发布日期:2022-03-26
  • 通讯作者: 冶继民 E-mail:jmye@mail.xidian.edu.cn

Reliability Analysis of Dependent Masked Data Systems with Burr Ⅻ Components

LI Yue, YE Jimin   

  1. School of Mathematics and Statistics, Xidian University, Xi’an 710126, China
  • Received:2021-06-28 Online:2022-03-26 Published:2022-03-26

摘要: 考虑相依串联屏蔽数据系统的可靠性分析问题. 首先, 在部件寿命服从Burr Ⅻ 分布的情形下, 通过引入Copula函数建立部件寿命变量之间的相依结构, 以及相依串联屏蔽数据系统的可靠性模型, 并推导出串联系统的一些概率结果; 其次, 基于逐步Ⅱ型截尾的系统失效数据, 得到模型参数的极大似然估计, 并基于渐近正态性理论和bootstrap抽样算法, 构造模型参数的渐近置信区间及偏差校正的百分位bootstrap置信区间; 最后, 进行仿真模拟和真实数据分析, 结果表明, 该模型方法对相依屏蔽数据系统的可靠性分析可行且有效.

关键词: Burr Ⅻ分布, Copula函数, 屏蔽数据, 串联系统, 偏差校正的百分位bootstrap方法

Abstract: We considered the problem of the reliability analysis of the dependent series system with masked data. Firstly, when the component life followed Burr Ⅻ distribution, the dependent structure of component life variables and the reliability model of dependent series system with masked data were established by introducing Copula function, and some probability results for the series system were derived. Secondly, Based on the progressively Type-Ⅱ censored system failure  data, we obtained the maximum likelihood estimation of model parameters, and the asymptotic confidence intervals and bias-corrected percentile bootstrap confidence intervals of model parameters were constructed based on asymptotic normality theory and bootstrap sampling algorithm. Finally, simulation and real data analysis were carried out. The results show that the proposed model  methods are feasible and effective for the reliability analysis of dependent masked data systems.

Key words: Burr Ⅻ distribution, Copula function, masked data, series system, bias-corrected percentile bootstrap method

中图分类号: 

  • O213.2