Research On Fault Diagnosis And Sensorless Control Of Five-Phase Permanent Magnet Synchronous Motor With Inter-Turn Short Circuit Impact

In order to take advantage of the strong fault tolerance for five-phase permanent magnet synchronous motor(PMSM),and improve the reliability of aerospace vehicle electric drive system,it is able to adopt the sensorless fault tolerant control method to ensure the motor still with the fault tolerant ability in fault state.In many electrical faults of motor,the inter-turn short circuit(ITSC)fault with higher probability declines the driving performance of the electric drive system rapidly and meanwhile the third harmonic generated by the star winding connection of five-phase PMSM seriously affects the sensorless control accuracy of the electric drive system.For the reason,it is necessary to diagnose the related fault information of the driving motor in time firstly and then take the sensorless control method with the consideration of the third harmonic effect as the basis for the subsequent fault tolerant control to ensure that the electric drive system is still with good fault tolerant operation capability.So the thesis focuses on the ITSC fault diagnosis and the sensorless control considering the third harmonic for five-phase PMSM and the main innovations are as follows:(1)A new stator winding ITSC fault model and a sensorless control model with the third harmonic effect are built for five-phase PMSM.By introducing ITSC ratio,the established ITSC fault model can accurately describe the different fault levels and fault location for one-phase and multi-phase winding ITSC faults of five-phase PMSM,and also can be combined with a variety of extreme optimization algorithm,which provides a model foundation for the ITSC fault diagnosis of five-phase PMSM.The established five-phase PMSM sensorless control model is considered with the effects of the third harmonic on the related physical parameters such as motor current,voltage and flux linkage,as well as the coordinate transformation matrix,which is more accurate than the conventional five-phase PMSM sensorless control model and meanwhile provides a model foundation for five-phase PMSM sensorless control with the third harmonic impact.(2)A fault diagnosis method based on Particle Swarm Optimization(PSO)is proposed for five-phase PMSM ITSC fault.The proposed method effectively solves the problem of premature convergence for conventional PSO algorithm and transforms the ITSC fault diagnosis into the extremum optimization of fault parameter,ITSC ratio,in the five-phase PMSM fault model,which simplifies the diagnosis and detection process of the ITSC fault with good optimization ability for ITSC fault parameter,and can diagnose and detect any one or two phase windings ITSC fault in different fault location and level for five-phase PMSM.(3)A fault diagnosis method based on trust region algorithm is proposed for five-phase PMSM ITSC fault.The proposed method combines BFGS quasi-Newton method with the conventional trust region algorithm,which not only stabilizes and globally converges,but also improves the convergence speed and computational efficiency of the extremal optimization algorithm effectively.Besides,the proposed method also transforms the ITSC fault diagnosis into the extremum optimization of fault parameter,ITSC ratio,in the five-phase PMSM fault model,which simplifies the diagnosis and detection process of the ITSC fault with good optimization ability for ITSC fault parameter.Furthermore,the proposed method not only can diagnose and detect any one or two phase windings ITSC fault in different fault level for five-phase PMSM,but also can realize the diagnosis and detection of five-phase PMSM with more than two phase windings ITSC fault in different fault level,and even if it is necessary to optimize each phase's ITSC ratio in the parameter model of fault motor,the method is still with faster convergence speed and higher computational efficiency.(4)An adaptive sliding mode observer(SMO)is proposed for sensorless control of five-phase PMSM with the third harmonic impact.The sliding mode current observer in the proposed method is with smaller chattering and can extract the back-electromotive force(EMF)equivalent signal more accurately.And the back-EMF adaptive observer designed as the proposed method eliminats the low-pass filter and phase compensation module which are necessary in the conventional SMO sensorless control system,so the structure of the SMO is simpler.Besides,the proposed method takes full consideration of the third harmonic effect on the estimation of the motor speed and rotor position in sensorless control of five-phase PMSM.The designed adaptive SMO is stable and convergent,and the estimation error of motor speed and rotor position is smaller than conventional SMO,which can be applied in middle and high speed applications with strong robustness,and still with good effect under ITSC fault state.(5)An adaptive parameter estimation method is proposed for sensorless control of five-phase PMSM with the third harmonic impact.The proposed method can estimate the motor speed and rotor position by only measuring one phase's voltage signal,current signal and differential current signal without the extraction of back-EMF equivalent signal and the filtering of high-order harmonics,which also omits the low-pass filter so the structure of the SMO is simpler than conventional SMO,and has less effect of the low signal-to-noise and the third harmonics on back EMF extraction and speed and rotor position estimation,whose steady-state estimation error is zero theoretically.Furthermore,the proposed method can be applied to the speed and rotor position estimation in wide speed range and the estimation error of motor speed and rotor position under ITSC fault state is smaller than SMO.The proposed method is not only suitable for five-phase PMSM sensorless control system,but also can be generalized and applied to the three-phase PMSM sensorless control system.The thesis establishes the ITSC fault model and sensorless control model with the third harmonic impact for five-phase PMSM and proposes two kinds of fault diagnosis methods based on PSO and trust region,and two kinds of sensorless control methods based on adaptive SMO and adaptive parameter estimator,which provides new approaches for fault diagnosis of five-phase PMSM with one-phase or multi-phase windings ITSC fault,and speed and rotor position estimation with the third harmonic impact in sensorless control of five-phase-PMSM,and also provides the foundation for sensorless fault tolerant control of five-phase PMSM with ITSC fault.