Research on Faulty Speed Sensor Identification in High-Speed Train Traction System Based on Electromagnetic Torque Correlation Analysis

In high-speed train traction systems, when a speed sensor malfunctions, it generates abnormal signals that propagate within the traction system, making it challenging to identify the faulty sensor. This paper focuses on the scenario of a single speed sensor failure and utilizes the Convergent Cross Mapping (CCM) algorithm to calculate the electromagnetic torque correlation between traction motors. Based on this analysis, a feature vector representing the fault patterns of speed sensors is constructed, and an electromagnetic torque correlation-based vector method (EMC-CVM) is proposed for speed sensor fault identification. Additionally, a simulation model of the main circuit of the traction system is developed using Matlab/Simulink. The simulation results under different speeds and load torque conditions demonstrate the effectiveness of the EMC-CVM method in accurately identifying faulty speed sensors in the traction system.