Sensor Fault detection and fault tolerant control of induction motor drivers for electric vehicles

An effective fault detection and fault tolerant control strategy in induction motor is described in this paper. Based on the algorithm of posterior reliability voting, the proposed approach could detect the healthy state of encoder, and adjust the weights between the speed from encoder and estimated speed based on MRAS. When redundant current sensors are installed in the controller, a logic decision scheme could detect and shield the faulty one, and guarantee that the current signals involved in the control are from the two healthy current sensors. Abrupt faults of both encoder and speed sensor are simulated in the induction motor driver individually, and simulation results in terms of speed response indicate the effectiveness of the proposed strategy subject to faulty conditions caused by encoder or current sensor.