A Multi-state Fault Tree Analysis Based Bayesian Network Modeling Method for Failure Prediction

摘要：

By exploring the practical cause-effect relationships embedded in the multi-state fault tree analysis (MFTA), this paper proposes a novel modeling method for failure prediction Bayesian Network (FPBN) to establish instance models and facilitate applications. First, the multi-state events and logic gates in MFTA are shifted into variables, edges and conditional probability distributions (CPD) in FPBN. Then, the detection variables are also modeled in FPBN by considering the effects of modern detection devices. At last, the full FPBN modeling approach based on MFTA is put forward to direct the modeling process.

关键词： Multi-state Fault tree analysis Failure prediction Bayesian network Modeling

作者: ZHIQIANG CAI SHUDONG SUN SHUBIN SI NING WANG

作者单位: Ministry of Education Key Laboratory of Contemporary Design and Integrated Manufacturing Technology, Northwestern Polytechnical University, Xian, 710072, P.R. China

会议类型: 国际会议

会议名称: The 7th Inyernational Conference on Mathematical Methods in Reliability:Theory,Methods,Applications(第七届国际可靠性数学、理论、方法与应用会议 MMR2011)

会议地点: 北京

会议语种:英文

页码: 314-319

在线出版日期: 2011-06-20（万方平台首次上网日期，不代表论文的发表时间）