Design Of JTC Fault Diagnosis System Based On The Train Rear Data Of The Detection Vehicle

As an important part of the train operation control system,jointless track circuit(JTC)plays an important role in train occupy checking and the continuous information transmission from the ground to the vehicle.At present,periodic detection of the detection vehicle is an important measure of JTC fault diagnosis.Consider that the train head track circuit reader(TCR)and the train rear TCR can work normally in actual operation,they can sense the track circuit signals of their respective positions and form the corresponding train head data and train rear data.From the basic structure and working principle of the JTC and the detection vehicle,it can be seen that the above two kinds of data have obvious differences in signal composition and intensity.Based on these differences,research on the JTC fault diagnosis method based on the train rear data to improve the JTC fault diagnosis level of the detection vehicle has important practical significance and research value.The main research work of this paper is as follows:1.According to the basic structure and working principle of JTC and the train rear TCR equipment of the detection vehicle,based on transmission line theory,the train rear TCR induced voltage amplitude envelope is modeled for the first time and its influencing factors are analyzed.A four-terminal network information transmission equivalent model of the train rear TCR induced voltage amplitude envelope is built.Based on the built model,the influence of parameters and equipment such as ballast resistance,compensation capacitor,tuning unit and SVA on the train rear data is analyzed through fault injection technology.2.A JTC fault diagnosis method based on one-dimensional convolutional neural network(CNN)is proposed to implement the diagnosis of the combined faults of the tuning area equipment and the estimation of ballast resistance.Based on the above research on influencing factors,CNN for fault diagnosis of equipment in the tuning zone and CNN for ballast resistance regression estimation are constructed.A network training set for different track circuit carrier frequencies,track lengths and ballast resistances is constructed,and the corresponding network training is carried out to improve the versatility and accuracy of the constructed neural network.3.Based on the above fault diagnosis algorithm,the corresponding JTC fault diagnosis software used for the detection vehicle rear is designed and implemented,and it is integrated with the signal detection system of the detection vehicles.According to on-site application requirements,the software interface design and function design were carried out,and each functional module was verified.The experimental results show that the fault diagnosis method of JTC tuning area equipment and ballast resistance proposed in this paper based on the train rear data has high accuracy,and it can meet the application requirements of the railway field.At the same time,it can also provides a new idea for the fault diagnosis of JTC based on the detection vehicle.This paper contains 64 figures,8 tables and 51 references.