基于改进Bootstrap-Bayes的电子式剩余电流动作断路器可靠性评估  

Reliability Evaluation of Electronic Residual Current Operated Circuit Breakers Based on Improved Bootstrap-Bayes

作　　者：刘帼巾[1] 王泽[1] 李想[1] 赵兴洲 缪建华 Liu Guojin;Wang Ze;Li Xiang;Zhao Xingzhou;Miao Jianhua(State Key Laboratory of Reliability and Intelligence of Electrical Equipment Hebei University of Technology,Tianjin 300130,China)

机构地区：[1]省部共建电工装备可靠性与智能化国家重点实验室(河北工业大学),天津300130

出　　处：《电工技术学报》2022年第16期4250-4258,共9页Transactions of China Electrotechnical Society

基　　金：河北省自然科学基金项目(E2020202221);河北省高等学校科学技术研究项目(ZD2019041)资助。

摘　　要：作为一种高可靠性、长寿命产品,电子式剩余电流动作断路器可获得的试验数据很少。为了对其可靠性进行评估,提出一种基于改进Bootstrap-Bayes的可靠性评估方法。首先,进行温度为加速应力的剩余电流动作断路器的加速退化试验,利用退化数据外推得到伪失效寿命,经检验,电子式剩余电流动作断路器伪失效寿命服从双参数威布尔分布;然后,采用灰色GM(1, 1)模型对试验数据进行扩充作为先验信息,通过马尔科夫链蒙特卡洛(MCMC)算法中的Gibbs抽样结合Bayes公式得到参数估计值;最后,利用Arrhenius加速模型实现正常使用环境下剩余电流动作断路器的可靠性评估。As a kind of high reliability and long-life product, the electronic residual current operated circuit breaker has few test data. In order to evaluate its reliability, a reliability evaluation method based on improved Bootstrap-Bayes was proposed. Firstly, with temperature as the accelerated stress, the accelerated degradation test for the residual current operated circuit breaker was carried out.The pseudo failure life was obtained by extrapolating the degradation data, and it was verified to obey the two-parameter Weibull distribution. Then, the GM(1, 1) model was used to expand the experimental data as the prior information, and the parameter estimates were obtained by Gibbs sampling in the Markov Chain Monte Carlo(MCMC) algorithm combined with Bayes formula. Finally, the Arrhenius acceleration model was used to evaluate the reliability of the residual current operated circuit breaker under normal use environment.

关 键 词：电子式剩余电流动作断路器 GM(1 1)模型 Bootstrap抽样 BAYES MCMC 可靠性评估 

分 类 号：TM506[电气工程—电器]