Research Of Track Circuit Fault Diagnosis Method Based On Comprehensive Signaling Monitoring Maintenance System

Railway comprehensive signaling monitoring maintenance system is the product of the development of science and technology and big data era,it can real-time monitor the working state of the railway signal equipment,ensure traffic safety,auxiliary troubleshooting,guide the field maintenance,and also can improve the efficiency and level of maintenance department.At present,our railway comprehensive signaling monitoring maintenance system mainly supervise the digital quantity and analog quantity of the signal device,provide the parameter of the equipment for the signal workers,and provide scientific basis for repair and analyse the facility.The alarm’s analysis and judgment mainly depend on the corresponding threshold value and logic analysis,it well be slightly insufficient in the intelligent alarm analysis and generalization performance,so we need take certain intelligent diagnosis method to provide a more reliable guarantee for equipment fault analysis and diagnosis.The research object of this thesis is the track circuit of railway comprehensive signaling monitoring maintenance system,the main work in the thesis are as follows:Firstly,this thesis introduces the existing railway comprehensive signaling monitoring maintenance system,analyze the alarm and early-warning of the track circuit subsystem in detaile,and extract several typical track circuit faults,also give the corresponding judgment condition.Then extracts alarm data in the original data according to the data protocol.Secondly,for the part of track circuit alarm of the comprehensive signaling monitoring maintenance system,because of different track circuit has different parameters and electrical properties,some field failure data has large fluctuation,so when we use algorithm to diagnosis the fault,it may has false positives or false negative.Therefore,the thesis use similarity analysis before diagnosis the fault with intelligent diagnosis,it can extract the fluctuate wildly data,and judge it by field comprehensive signaling monitoring maintenance system diagnostic method;less volatile data can use GA-BP algorithm for the diagnosis.The experimental results show that the method can partly solve the false positives or false negative problem when we use algorithm to diagnosis the fluctuate wildly data,and it can also accelerate the convergence rate,improve the algorithm of fault diagnosis precision,has a certain practicability.Then,for the part of track circuit early-warning of the comprehensive signaling monitoring maintenance system,The thesis uses the multi-level fuzzy comprehensive evaluation method to divide the level of the track circuit warning system.According to the analysis of track circuit warning system,select the appropriate warning indexes,set up warning index threshold and multi-level fuzzy comprehensive evaluation model,also build evaluation function and membership function to each impact factor to comprehensive judge the track circuit warning system.For the track circuit voltage’s abnormal and fluctuation warning,according to the field experience,use interpolation method to the evaluation of single factor.Through the data calculation,verify the feasibility of the fuzzy comprehensive evaluation model.The model both considered the effects of railway voltage’s overrun,abnormal,fluctuation and the external environment factors,improve the safety and reliability of the track circuit warning system,it is Convenient for electricity workers to timely and fast to deal with early-warning information.Finally,this thesis summarizes the research contents,and put forward some shortages,point out some problems in the comprehensive signaling monitoring maintenance system,and then points out further research work.