Aiming at the problem of the major safety risks for the train operation caused by the fault of the current sensor in the subway train traction system,a new fault monitoring and tolerant control algorithm was proposed.Firstly,a current sliding mode observer was designed by using the static mathematical mode of the three-phase asynchronous motor and the specific relationship combined the rotor flux and the counter electromotive force of the motor.Secondly,the current residual was generated by comparing the measured value of the current sensor with the current estimation in the current sliding mode observer,and the working state of the current sensor was monitored and located by the modified Bayes classification algorithm in real time.When the fault occurred,the system was controlled by the current reconstruction value instead of the fault value in the original current sensor.Finally,the feasibility of the scheme was verified by the MATLAB/Simulink simulation platform,which was compared with other three typical fault-tolerant control algorithms.The research results show that the scheme has higher accuracy and stronger robustness,which is based on the counter electromotive theory of the three-phase asynchronous motor.In terms of three-phase current reconstruction,the comprehensive error is reduced by 24.2%,which has better control effect.