摘要
针对列控中心系统中因设备故障信息不完善、失效机理不明确等因素产生的认知不确定性,以及动态失效、共因失效、恢复机制问题,提出了将证据理论和动态贝叶斯网络相结合的动态证据网络方法,并对列控中心系统进行了可靠性评估及重要度影响分析。在分析系统各单元功能逻辑关系的基础上建立故障树,并将故障树转化为动态贝叶斯网络。通过动态贝叶斯网络正向推理,并结合信任测度和似然测度得到了列控中心系统的可靠度区间以及讨论共因失效对系统可靠性的影响。此外,通过动态贝叶斯网络反向推理得到了列控中心系统薄弱环节。最后,通过求解重要度的大小,探究底事件可靠度变化和认知不确定性对系统可靠度的影响程度。结果表明:动态证据网络能够较好地处理系统可靠性评估中的不确定信息。相比于在完全信息条件下,运用动态证据网络方法得到的可靠性评估结果较为合理,也更加符合实际。
The train control center(TCC)system has epistemic uncertainty caused by imperfect equipment failure data and unclear failure mechanisms,dynamic failure,common cause failure,and recovery mechanism.This paper presents a systemic approach to evaluate the reliability of the TCC system by combining the evidence theory with a dynamic Bayesian network.Firstly,taking account of the functional logic relationship of system units,a fault tree was established and transformed into a dynamic Bayesian network.Secondly,by using the forward inference ability of a dynamic Bayesian network,the reliability of the TCC system was obtained by combining the belief measure with the plausibility measure of the evidence theory.The effect of common cause failure on system availability was discussed.Using the backward inference ability of a dynamic Bayesian network,the weaknesses of the TCC system were recognized.Finally,by solving the magnitude of Birnbaum and cognitive importance,the influence of basic units with reliability changes and the epistemic uncertainty on system reliability was explored.The results show that the reliability of the TCC system with common cause failure at Week 50 is[0.8170,0.7599].The reliability of the TCC system without common cause failure at Week 50 is[0.8403,0.7897].It can be found that the common cause failure has an impact on the reliability of the TCC system,which will lead to a decrease in the reliability of the system The Port Input/Output unit and the Communication Interface Unit are weak links with system reliability.To further improve the reliability of the TCC system and formulate a planned maintenance strategy,they should be paid more attention.This method enhances the ability of dynamic Bayesian networks to deal with uncertain information.It can deal with the epistemic uncertainty in system reliability evaluation.Compared with the condition of complete information,the reliability evaluation results of the TCC system obtained through this method are more reasonable and more realistic.
作者
朱爱红
董国庆
ZHU Aihong;DONG Guoqing(School of Automation and Electrical Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)
出处
《安全与环境学报》
CAS
CSCD
北大核心
2023年第9期2993-3001,共9页
Journal of Safety and Environment
基金
国家自然科学基金项目(52162050)
中国国家铁路集团有限公司科技研究开发计划系统性重大项目(P2021T003)。
关键词
安全工程
列控中心
认知不确定性
证据理论
重要度
可靠性评估
safety engineering
train control center
the epistemic uncertainty
evidence theory
importance
reliability evaluation
