Current Sensor Fault Detection And Fault-tolerant Control Strategy For Speed-sensorless Motor Drive System Of Urban Rail Train

With the accelerated urbanization process,urban rail transportation has been developed rapidly and taken up a large amount of passenger traffic,to ease the pressure of urban population travel.At present,China’s urban rail transit operating mileage has ranked among the top in the world.While promoting economic development,it is vital to ensure the safe and reliable operation of urban rail trains.As the core equipment of urban rail trains,the reliable operation of the traction drive system is the basis and guarantee for the safe development of urban rail transportation.In the traction drive system of urban rail trains,vector control is widely used for its excellent performance and ease of implementation.Usually,vector control requires at least one speed sensor and two-phase current sensors.Once a sensor fails,the performance of vector control decreases significantly,even leading to safety accidents.To improve the reliability of the traction drive system of urban rail trains,this paper takes the induction motor applied in the traction drive system of urban rail as the research object,and investigates a current sensor fault-tolerant control algorithm for the speed-sensorless induction motor system.Firstly,a speed identification algorithm based on sliding mode observer(SMO)is designed for the induction motor,and the estimated speed is used to replace the measured value,realizing speed-sensorless control of the induction motor.Secondly,the mathematical model of common fault types of current sensor is established and the fault propagation mechanism and adverse effects of different fault types on the system are analyzed.Then,for different fault types,a method based on coordinate transformation is studied to realize the fault detection of current sensor under speed-sensorless induction motor drive system.On this basis,a second-order generalized integrator-frequency-locked loop(SOGI-FLL)is designed to reconstruct faulty phase current information.After detecting the current sensor failures,the reconstructed current information is used to maintain the estimated performance of the SMO for the motor speed and current.In addition,considering the possible DC offset in the sensor measurement,a SOGI-FLL scheme with DC bias suppression is investigated to further improve the reliability of the fault-tolerant control system.Finally,the correctness and effectiveness of the investigated current sensor fault-tolerant control scheme is verified by a low power motor experimental platform.This paper first implements the SMO-based speed-sensorless control of the induction motor,on the basis of which the current sensor fault detection and fault-tolerant control of the speed-sensorless control system are further verified.