基于贝叶斯网络的高铁系统概率安全评估方法研究  

RESEARCH ON PROBABILISTIC SAFETY ASSESSMENT IN RAILWAY SYSTEM BASED ON BAYES NETWORKS

作　　者：朱海宏[1] 徐中伟[1] 梅萌[1] 祝玉军[1] 

机构地区：[1]同济大学信息与通信工程系,上海201804

出　　处：《计算机应用与软件》2011年第3期26-28,31,共4页Computer Applications and Software

基　　金：国家自然科学基金项目(60674004);铁道部科技研究开发计划基金项目(2007X003)

摘　　要：高速铁路控制系统的安全性评估非常重要,引入贝叶斯网络技术,充分利用传统事件树、故障树的分析优势,将事件树中各安全环节的故障树分别转化为贝叶斯网络,并按逻辑关系最终融合为一张完整的贝叶斯网。通过整合的贝叶斯网络不仅可以分析列控系统的安全性,同时还能得出其他有用的概率推理信息。It is of great importance for the safety assessment of high speed railway control system.In this paper,the Bayes network technology is introduced which takes advantage of traditional event tree and fault tree that are good at analysis.The fault trees at each safety stage of the event tree are turned into Bayes network respectively,then are merged into an integrated Bayes network by logical relations.Through the integrated Bayes network,not only can the safety of the railway control system be analyzed,but also other useful probabilistic reasoning information can be obtained.

关 键 词：概率安全评估 安全性 事件树 故障树 贝叶斯网络 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]