列车控制中心系统多态可靠性分析OA

Due to the dynamic and polymorphic characteristics of fault modes in train control center,traditional continuous-time Bayesian network fails to effectively address the the issue of polymorphism.Moreover,missing fault data in practical engineering makes it difficult to obtain accurate fault data.To overcome these challenges,this paper proposed a reliability analysis method integrating hyper-ellipsoidal T-S fault tree with Bayesian network.Firstly,the hyper-ellipsoid model was employed to constrain the probability interval of bottom events,in order to address the issue of data uncertainty.Secondly,a Bayesian network model was constructed based on the T-S fault tree,and the fuzzy numbers were used to characterize multiple fault states of the nodes.Finally,the proposed method was used to analyze the reliability of the train control center system.Through forward reasoning,the probability curves for each state during the system operation were obtained.Through posterior probability analysis,the vulnerable components of the system were identified.The research results demonstrated that compared with the conventional interval T-S fault tree,the analysis accuracy and reasoning capability of the proposed method had been improved.Additionally,the forward reasoning and posterior probability analysis results can provide theoretical support for the maintenance and reliability optimization of the train control center.