基于动态贝叶斯网络的CTCS-3级ATP系统可靠性分析

Reliability Analysis on ATP System of CTCS-3 Based on Dynamic Bayesian Network

针对传统可靠性分析方法在分析CTCS-3级ATP系统可靠性时存在的不足,采用动态贝叶斯网络模型对其进行可靠性分析。首先,根据CTCS-3级ATP系统的结构与功能建立其故障树模型,并将故障树模型转化为相应的动态贝叶斯网络模型。然后,综合考虑维修、共因失效等因素,对ATP系统进行可靠性分析。结果表明:利用动态贝叶斯网络双向推理的优势对ATP系统进行可靠性分析,不仅可以计算出系统的可靠度,而且可以有效地识别系统的薄弱环节;在可靠性分析过程中,如果忽略共因失效和维修因素对ATP系统可靠性的影响,将导致分析结果偏差过大。通过诊断推理得到,除双系冷备单元外,列车接口单元是ATP系统的薄弱环节,因此加强对其维护检查能够有效降低事故的发生概率。

In response to the shortcomings of traditional reliability analysis method in the analysis of the reliability of the ATP system of CTCS-3,the reliability analysis on the ATP system was carried out by using the dynamic Bayesian network model.Firstly,the fault tree of ATP system of CTCS-3was constructed on the basis of its structure and function,before the fault tree was turned into a corresponding dynamic Bayesian network.Secondly,by considering the factors such as maintenance and common cause failure,reliability analysis of ATP system was made.The results showed that,by using the bidirectional inference of dynamic Bayesian network,the reliability of the ATP system of CTCS-3can be calculated,and the weak links of the ATP system can be effectively identified.Neglect of the impact of maintenance and common cause failure on the reliability of the system during reliability analysis will lead to a large deviation of the analysis results.The analysis of diagnosis reasoning showed that train interface unit is a weak link of the system apart from the double cold standby unit.Therefore,strengthening maintenance check of the train interface unit is helpful to reduce the failure rate of the system.