China Mechanical Engineering

Previous Articles     Next Articles

Reliability Analysis of Train Control Center Based on Discrete-time Bayesian Network

WANG Kang1,2;QI Jinping1,2,3;ZHOU Yahui1,2;LI Shaoxiong1,2;ZHAO Ruihu4;GUO Hao4   

  1. 1. Mechatronics T&R Institute, Lanzhou Jiaotong University, Lanzhou,730070
    2. Gansu Provincial Engineering Technology Center for Informatization of Logistics & Transport Equipment, Lanzhou,730070
    3. Gansu Provincial Industry Technology Center of Logistics&Transport Equipment, Lanzhou,730070
    4. China Railway Lanzhou Group Co., Ltd., Lanzhou,730000
  • Online:2021-02-25 Published:2021-03-05

[轨道交通运维技术]基于离散时间贝叶斯网络的列控中心可靠性分析

王康1,2;齐金平1,2,3;周亚辉1,2;李少雄1,2;赵睿虎4;郭浩4   

  1. 1. 兰州交通大学机电技术研究所,兰州,730070
    2. 甘肃省物流及运输装备信息化工程技术研究中心,兰州,730070
    3. 甘肃省物流与运输装备行业技术中心,兰州,730070
    4. 中国铁路兰州局集团有限公司,兰州,730000
  • 基金资助:
    国家自然科学基金(71861021);
    铁路总公司科研计划(2015T002-D);
    甘肃省高等学校科研项目(2018A-026);
    甘肃省重点研发计划(17YF1FA122);
    甘肃省高等学校科研项目(2018C-10)

Abstract: Aiming at the problems that the faults of electric multiple units train control center were both polymorphism and dynamic in actual working environment, an analysis method of discrete-time Bayesian network was proposed based on the functional logic relationship of each unit in the train control center. The multi-state characteristics of train control center fault were described by summarizing various fault modes of components, and the condition probability table was optimized by EM algorithm. Aiming at the dynamic failure of train control center, a dynamic Bayesian network model was established, then a task was divided into three stages: start, operation and braking. In each stage, the reliability of the model was analyzed by importance and sensitivity. Finally, the train control center of the CTCS-2 train control system was taken as an example to verify and analyze the discrete-time Bayesian network model, proving that the method may characterize the polymorphism and dynamic characteristics of train control center well.

Key words: train control center, reliability analysis, discrete-time Bayesian network, EM algorithm, importance

摘要: 针对动车组列控中心在实际工作环境中的故障同时具有多态性和动态性的问题,提出一种依据列控中心各单元的功能逻辑关系来建立离散时间贝叶斯网络的分析方法。归纳部件的多种故障模式并描述列控中心故障的多态特性,采用EM算法优化更新条件概率表;针对列控中心动态失效问题,建立动态贝叶斯网络模型,将一次任务划分为启动、运行、制动三个阶段,在各个阶段通过重要度和敏感性对该模型进行可靠性分析。最后,以CTCS-2级列控系统的列控中心为例,对该离散时间贝叶斯网络模型进行验证和分析,结果表明该方法能够很好地表征列控中心的多态性和动态性。

关键词: 列控中心, 可靠性分析, 离散时间贝叶斯网络, EM算法, 重要度

CLC Number: