Research On Fault Diagnosis Of High-speed Railway Traffic Dispatching System Based On Bayesian Network

With the rapid development of China's high-speed railways and the vigorous implementation of globalization strategy of China's high-speed railway and the strategy of building a country with strong transportation network,the requirements for safe and orderly operation of high-speed railways have been greatly improved.This paper,takes the high-speed railway traffic dispatching system as the research object,uses the theory and model of Bayesian network to conduct qualitative and quantitative analysis of the fault diagnosis of the high-speed railway traffic dispatching system,which is of great significance to the safe operation of China's high-speed railway.The main contents are as follows:1.The fault characteristics of the high-speed railway traffic dispatching system are analyzed in detail,based on the actual fault case and expert consultation.Secondly,the fault characteristics of the high-speed railway traffic dispatching system are divided into fault sources coupling and fault performance uncertainty,meanwhile the fault evolution process of the high-speed railway traffic dispatching system is constructed.2.A discriminating system for fault sources of high-speed railway traffic dispatching system is proposed,the fault sources of the high-speed railway traffic dispatching system are comprehensively analyzed,based on the relevant standards,legal and regulatory documents and the actual fault case.The gray correlation analysis method is used to quantitatively calculate the degree of correlation between the fault sources and the fault of the high-speed railway dispatching system.3.This paper takes the human fault sources with the highest degree of association with the failure of the high-speed railway dispatching system and the most common temporary speed limit as an example.Firstly,the fault diagnosis process of Bayesian network and the fault diagnosis Bayesian network layered model is established.Secondly,the Bayesian network aggregation algorithm combined with expert consultation is used to determine the prior probability of Bayesian network nodes.Finally,the Bayesian network based on expert knowledge and the Bayesian network based on structure learning algorithm are combined,which determines the optimal Bayesian network for fault diagnosis of high-speed railway traffic dispatching system.4.The optimal Bayesian network for fault diagnosis of high-speed railway traffic dispatching system is constructed in GeNIe2.1 and the probability of failure of the high-speed railway traffic dispatching system and the probability of each fault source in the event of failure are calculated through causal inference and diagnostic inference.Thereby the purpose of fault diagnosis of high-speed railway traffic dispatching system is achieved,which will provide strong support for quickly and accurately determining the hazard and eliminating high-speed railway dispatching system faults.