Reliability Analysis of Fuzzy Bayesian Networks Based on Uncertain Ordered Weighted Operators

Abstract

After analyzing the shortcomings of traditional fault tree analysis methods, a fuzzy Bayesian network reliability analysis method based on fault tree is proposed. This method of modeling uses the Bayesian method, the event polymorphism of complex systems is described by the node polymorphism expression feature of Bayesian network theory, and the uncertain logical relationship between events is described by the conditional probability table of Bayesian network. Based on the Bayesian model, the fuzzy set theory is introduced, and the experts fuzzy evaluation of event probability is described by triangular fuzzy numbers. In the evaluation information of the experts with uncertain weights, the expert evaluation information of the uncertain weights is calculated by using the uncertainty-ordered weighted average operator to calculate the expert weights, and finally the exact value of the occurrence probability of different states is obtained. Substituting it into the Bayesian network to calculate the probability of occurrence of different states of the leaf nodes, and then calculating the posterior probability of each root node and its importance.