Asynchronous Motor Fault Diagnosis Based On Multiple Signal Decomposition And Reconstruction

Traction motor usually using three-phase cage asynchronous motors is the main component of high-speed railway traction drive system,which stable operation plays a vital role in the safe operation of railway.However,under complex operating conditions,asynchronous motors are prone to fault.If the faults of motor are not eliminated in time,the operating motor with a fault will seriously threaten the safe operation of railway.Therefore,it is of great significance for ensuring the safe operation of railway to develop the research of asynchronous motor fault diagnosis technology,realize the fault identification and diagnosis in time,and take effective countermeasures to prevent the development of fault.The research work on the fault diagnosis method for asynchronous motors in this paper is as follows:Firstly,the failure mechanism of the stator,rotor and bearing fault of asynchronous motors is analyzed respectively,and the fault characteristic signals of the stator current of stator and rotor fault and the vibration signal of the bearing fault are obtained.Secondly,according to the equivalent physical model of stator interturn short circuit fault and broken rotor bar fault of asynchronous motors,mathematical model of stator and rotor fault are analyzed,and then the fault models are built in Matlab/Simulink.The validity of the built models is verified by detecting corresponding stator current fault feature signals.Thirdly,the DC-link current is selected as the monitoring signal for the stator interturn short circuit fault and the broken rotor bar fault,and the stator and rotor fault DC-link current expressions are obtained according to the relationship among the inverter switching function,the input and the output current.According to the characteristics of the DC-link current,a diagnosis method of stator and rotor fault based on multiple signal decomposition and reconstruction is explored;According to the characteristics of vibration signal,a diagnosis method of bearing fault based on multiple signal decomposition and reconstruction is explored.Finally,on the hardware-in-the-loop test platform composed of dSPACE and TMS320F28335,the asynchronous motor fault model of stator-turn short-circuit fault and broken rotor bar fault are built,and the above fault diagnosis method of stator and rotor fault is experimentally verified.The fault identification and diagnosis results verify the reliability and practicability of the stator and rotor fault diagnosis method.The rolling bearing fault data of the Case Western Reserve University laboratory in the United States is used to experimentally verify the bearing fault diagnosis method explored in this paper.This method can extract fault characteristic signals from complex vibration signals to realize the identification and diagnosis of bearing inner ring and outer ring fault.The reliability and practicability of bearing fault diagnosis method is verified.