Research On Open Circuit Fault Diagnosis Of Six-phase Permanent Magnet Synchronous Motor Inverter

In recent years,more and more attention has been paid to fault diagnosis and fault tolerance of multi-phase permanent magnet synchronous motor(PMSM),and the accuracy of fault signal diagnosis and fault prediction processing has become the key technology to eliminate motor fault.The fault information is of great help to the fault tolerant control of the motor,so it is of great significance to diagnose the fault information timely and accurately.Based on the project of national natural science foundation of China "fault diagnosis and fault-tolerant control research of the control system driven by multi-phase motor for electric vehicles(61603263)",this project conducts fault-tolerant control research on the faults of the one-phase open circuit and two-phase open circuit of the six-phase permanent magnet synchronous motor.Firstly,this paper takes the six-phase permanent magnet synchronous motor inverter as the research object,obtains the mathematical model of the inverter circuit by using the switching function,and builds the six-phase permanent magnet synchronous motor drive system model on the MATLAB/Simulink platform.The open circuit faults of single bridge arm and double bridge arm of inverter are divided into four categories,and the variation of the electrical output of the motor is discussed with an example of each type of fault,and the stator current of the motor is selected as the research object to extract the fault characteristic.Then,aiming at the problem of white noise interference in the collected signals,an adaptive RLS Volterra noise cancellation system was selected to improve the EMD method.The EMD energy entropy and improved EMD energy entropy fault diagnosis simulation analysis were carried out in each type of fault.By comparing the simulation results of the two fault diagnosis methods,it can be concluded that the improved EMD energy entropy fault diagnosis method based on adaptive RLS Volterra filter can effectively eliminate the interference problem of white noise and has higher effectiveness.In this paper,an improved EMD energy entropy and normalized average current method are used to diagnose single and double bridge open circuit faults of the six-phase permanent magnet synchronous motor(PMSM)inverter.The threshold diagnosis system was established in the simulation software,and the method was verified for the normal operation state and four kinds of fault states respectively.The threshold diagnosis system was established in the simulation software to verify the methods of normal operation state and four types of fault state respectively.The simulation results show that the method adopted in this paper is effective to a certain extent,and is not affected by load changes,and has a certain anti-noise interference ability.Finally,a method based on the combination of fault characteristic quantity and support vector machine is proposed to study the open circuit fault diagnosis of six-phase permanent magnet synchronous motor inverter.Aiming at the open circuit faults of single and double bridge arms of inverter,six phase current fault feature vectors were constructed by using improved EMD energy entropy and normalized average current method,and a pair of multitype support vector machines were selected to train and test the fault samples.Through the analysis of diagnosis accuracy and sample training,it is concluded that the fault diagnosis method based on improved EMD energy entropy and normalized mean current method combined with PMM classification has better accuracy and effectiveness.