Research And Design On Fault Diagnosis System For25Hz Phase Detecting Track Circuit

Most of the track circuit equipments are arranged in outdoor, they work in harsh andcomplex environment for long term, and then it’s very common phenomenon that equipmentscan’t run smoothly or even are scrapped. Especially for the station that ballast bed is pollutedor automatic drainage performance is poor, when it confronts strong rainfall or wet weatherseasons, track circuit is likely to occur over large leakage current, it may result in abnormalvoltage track circuit diagnosis, namely red-light-strap or poor shunting phenomenon, it willbring about considerable harmful effects on passenger transportation, and will destroy the safeoperation of trains, even will come to serious accidents. In addition, many stations are locatedin remote suburbs, so transportation condition is harsh and inconvenient, maintenance staffcan only get to fault location on foot, so achieving the regular maintenance of track circuit isalways a challenge for the railway electrical service sectors. Therefore, it is necessary todevelop a track circuit fault diagnosis system for solving the actual site’s problem, thepurpose is to eliminate track circuit failure in the bud timely, and guarantee safe operation oftrains.The thesis mainly establishes track circuit fault diagnosis system for the station25Hzphase detecting track circuit, and realizes the purpose that the diagnosis of red-light-strapfailure and poor shunting that are caused by environmental factors malfunction are diagnosedand alarmed.Firstly,25Hz phase detecting track circuit is selected as the research object, its featuresare introduced, thought analyzing the factors that affect track circuit red-light-strap and poorshunting, then the main factor is selected as a key indicator of its diagnosis. Combined withsite conditions, the warming deadlines are defined according to the track voltage monitoringgraph from the microcomputer monitoring. Three types of track circuit equivalent links’four-terminal network models are established, analysis software for25Hz phase detectingtrack circuit system is compiled by using MATLAB programming techniques, that is easy todetermine the warming deadlines.Secondly, the overall structural of25Hz phase detecting track circuit fault diagnosissystem is designed, the fuzzy theory and BP neural network algorithm are selected to combineintelligent analysis and processing methods, track circuit fault diagnosis based on fuzzy BPneural network model is established, the correctness and accuracy of the model are verified.Finally, fault diagnosis system software is completed.25Hz phase track circuit faultdiagnosis system of man-machine interface is complied by using Visual C++6.0developmentenvironments with Access database. The debugging results show that the system can give acorresponding early warning signal timely and quickly when the track circuit is performed to have a tendency to failure, and notify the Signal&Telecommunication people to take somemeasures to eliminate the faults. The system not only provides condition for25Hz phasedetecting track circuit fault diagnosis, provides a basis for its daily maintenance and troubleshooting, but also lays the foundation conditions for completing track circuit healthmanagement system in the future.