Application Research On Fault Diagnosis Of On-board Equipment Of Train Control System Based On Fuzzy Theory And Bayesian Network

Chinese train control system level 3(CTCS-3) is one of the important technologies in intelligent railway signal. The main function of CTCS-3 is to command train operation safely in order to improve the efficiency of railway transportation and its working conditions.With the rapid development of high-speed railway, fail-safe problems in train control system are becoming more and more important, especially the failure rate and performance of on-board equipment in train control system are more outstanding in the whole system operation. Because bayesian network fault analysis method has good causality and integrity, it was widely used in system diagnostic analysis. Since apriori information is difficult to determine, single bayesian network is difficult in the complicated system. In this thesis, CTCS3-300 T on-board equipment in train control system was taken as the research object, based on fuzzy mathematics and aknowledge of mathematical statistics, a method is put forward to solve fault type diagnosis analysis comprehensive bayesian method of on-board equipment system.In this thesis, from the viewpoint of systems science, the principle of CTCS3-300 T on-board equipment structure are analyzed, and the result in the failure of hardware and software are discussed in this thesis. Combined with the operating line fault tree analysis, fault statistics of bayesian network model is set up, and verify the diagnosis algorithm of the systems, the concrete research content as follows:(1) The analysis and research of CTCS3-300 T on-board system principle diagram and structure diagram, known as to its working principle and operating mode. Using the fault tree and fault type impact analysis method of fault analysis to set up bayesian model. Using bayesian network toolbox to calculate system simulation model based on single bayesian and fuzzy bayesian comparison of the two methods, reasonable fuzzy bayesian decision method is put forward.(2) The bayesian decision theory is analyzed, fuzzy theory is applied to the bayesian decision making in the apriori information, sample information and posterior information. Then the fuzzification of sample information and apriori information in the bayesian network and using the theory of fuzzy composite intersection of the adjusted fuzzy single information, combined with the point of failure loss index, calculated the fuzzy posterior information, according to the decision theory of loss function method to calculate the loss function of the expected to make decisions.(3) Combined with the three years of Wuhan-Guangzhou high-speed train CTCS3-300 T on-board equipment failure repair service statistical analysis of the structure and principle, the fuzzy bayesian decision method is applied to the instance. The results showed that the established decision model for on-board equipment troubleshooting process takes aspects have sharply reduced and the CTCS3-300 T on-board equipment security weaknesses existing in the research to provide certain reference.