Bayes Monte-Carlo Assessment Method of Protection Systems Reliability Based on Small Failure Sample Data

Abstract

Reliability assessment methods that need large failure sample sets would face a bottleneck while the “high reliability” of the protection system is taken into account and would, therefore, directly affect the credibility of the assessment results. To overcome this inadequacy, a protection system reliability assessment method based on small failure samples was proposed. First, after analyzing the operating characteristics of protection devices and its small failure sample data, the Weibull distribution was selected as the failure distribution through the goodness-of-fit test. Second, the parameters of the Weibull distribution were estimated by regression analysis, and Monte-Carlo simulation was used to determine the prior distribution of the Bayes estimation. The posterior probability and reliability indices were then obtained by using the Bayes theory. The case study demonstrates the effectiveness of the method.