Reliability analysis of CTCS-1 ATP based on enhanced Bayesian network

LIU Zhongtian; XU Yue; WANG Hao

doi:10.11860/j.issn.1673-0291.2017.05.009

Abstract：China Railway Corporation has proposed the development requirements of CTCS-1 train control system based on design and application experience of CTCS-2 and CTCS-3 train control system.The vehicle ATP equipment in this system will replace LKJ and cab signal controlling the existing railway line.CTCS-1 ATP equipment is studied in this paper and comparative analysis of reliability of CTCS-1 ATP system is conducted between modular and system redundancy.Firstly,CTCS-1 ATP reliability block diagrams are established according to the integral technical scheme of CTCS-1 train control system.Correspondingly,the enhanced Bayesian network model is set up.Secondly,the quantitative reliability indexes of CTCS-1 ATP on the foundation of modular and system redundancy are obtained through forward reasoning of enhanced Bayesian network,then the advantages and disadvantages of these two designs are evaluated.Finally,the basic unit conditional failure probability of ATP equipment is obtained through the backward reasoning of enhanced Bayesian network,and the potential weak links of two redundancy designs are found out.The evaluation results show that the RAM indexes of CTCS-1 ATP under the condition of two redundancy designs can meet the requirements of Chinese train control system requirement specifications,but the RAM indicators of CTCS-1 ATP applying the modular redundancy design are better than those using the system redundancy design;and the conditional failure rate of peripheral cold standby module is higher than other modules,which is the weak link of CTCS-1 ATP.