{"title":"Time-Dependent Reliability Analysis of System Based on Dynamic Bayesian Fault Network","authors":"Yunwen Feng, Z. Song, Cheng Lu, Chuxiong Yin","doi":"10.1109/SRSE54209.2021.00061","DOIUrl":null,"url":null,"abstract":"Prior and sample data information are two key information for Bayesian model to analyze and predict accurately. In order to make the time-dependent analysis results accurately describe the trend of reliability degradation with time, a Dynamic Bayesian fault network model (DBFN) is constructed. Firstly, the prior information is combined with exponential, uniform and normal probability density distribution functions to calculate the failure rate of the system based on curve fitting. Secondly, the backwards reasoning of Bayesian network is used to realize reliability analysis, which can trace the fault phenomenon to the fault cause. Finally, the time-dependent sensitivity analysis is carried out and the trend with time is given. Using the Cabin Door indication system failure as a case, the results show that the failure rate calculated by the dynamic method is closer to the time-varying state of the system than the static value. The method provides an objective means for system time-dependent reliability analysis.","PeriodicalId":168429,"journal":{"name":"2021 3rd International Conference on System Reliability and Safety Engineering (SRSE)","volume":"557 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2021-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 3rd International Conference on System Reliability and Safety Engineering (SRSE)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/SRSE54209.2021.00061","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Prior and sample data information are two key information for Bayesian model to analyze and predict accurately. In order to make the time-dependent analysis results accurately describe the trend of reliability degradation with time, a Dynamic Bayesian fault network model (DBFN) is constructed. Firstly, the prior information is combined with exponential, uniform and normal probability density distribution functions to calculate the failure rate of the system based on curve fitting. Secondly, the backwards reasoning of Bayesian network is used to realize reliability analysis, which can trace the fault phenomenon to the fault cause. Finally, the time-dependent sensitivity analysis is carried out and the trend with time is given. Using the Cabin Door indication system failure as a case, the results show that the failure rate calculated by the dynamic method is closer to the time-varying state of the system than the static value. The method provides an objective means for system time-dependent reliability analysis.