Research On Key Technologies Of Permanent Magnet Synchronous Motor Drive System Based On Active Disturbance Rejection Control

Permanent magnet synchronous motors(PMSMs)have been widely applied in modern ac drive system with the advantages of simple structure,high power density,high reliability and convenient maintenance,especially in the fields with demanding motor performance and control accuracy such as robots,aerospace,numerical control machines,etc.Nevertheless,the PMSM is a typical nonlinear multivariable and strong coupling system,which is influenced by external disturbance,parameter perturbation and magnetic field nonlinearity,etc.Therefore,the PMSM control system requires to be further optimized in robustness and anti-disturbance ability.The active disturbance rejection control(ADRC)is independent of the accurate model of the controlled object,the extended state observer(ESO)of which can realize online disturbance estimation and has excellent anti-disturbance ability.Hence,the ADRC has attracted extensive attention of scholars at home and abroad.In order to improve the performance of the PMSM drive system in robustness and dynamic anti-disturbance ability,this dissertation studies the speed control,current control and sensorless control of the PMSM.In addition,the current harmonic suppression strategy of the PMSM is also studied to further optimize the speed control,current control and sensorless control of the PMSM.Firstly,in order to improve the robustness of speed control for the PMSM and bring into full play the superiority of the linear ADRC(LADRC)and nonlinear ADRC(NLADRC),a novel PMSM control method based on the linear/nonlinear ADRC switching control(SADRC)is presented in the dissertation.The SADRC possesses the LADRC feature of fast response with large error and the NLADRC feature of large gain with small error,which can ensure the estimation and compensation ability of total disturbance under different working conditions,and effectively improve the robustness and control accuracy of the system.Taking the nonlinear mechanism of the PMSM into account,the SADRC controller suitable for the PMSM is constructed,the anti-interference mechanism of which is analyzed in depth.The influences of each parameter on the dynamic and steady-state performance of the whole system are emphatically compared and studied with the help of frequency domain method and simulation tests,and then the physical significance and setting skill of the SADRC parameters are summarized.By the comparative experiments with the proportional integral(PI)control,LADRC and NLADRC strategies,the feasibility and effectiveness of the SADRC are verified.Secondly,in order to enhance the current performances,robustness and prediction accuracy of the PMSM,a switching-active-disturbance-rejection-based deadbeat predictive current control(SADR-DPCC)is proposed in the dissertation for the design of the PMSM current controller.The SADR-DPCC integrates a DPCC with a switching ESO(SESO),which is the core component of the SADRC.Parameter mismatches and external disturbances of the PMSM are taken as the estimated total disturbances,which is substituted into the predictive current model of the PMSM and observed in real time by the SESO to enhance the current prediction accuracy.The SESO combines the linear ESO(LESO)and nonlinear ESO(NLESO)via appropriate switching strategy,which not only possesses the merits of LESO and NLESO,but also overcomes the shortcomings of them.Furthermore,the proposed switching strategy simplifies the parameter tuning and stability analysis of the NLESO.By the comparative experiments with the traditional DPCC and active-disturbance-rejection-based DPCC strategies,the superiority of the SADR-DPCC method in robustness and control accuracy are verified.Thirdly,in order to improve the robustness and adaptability of sensorless control for the PMSM,a sensorless control of the PMSM based on the fuzzy active disturbance rejection observer(FADRO)is proposed.Since the back electromotive force(EMF)can be regarded as the estimated current disturbance,the active disturbance rejection observer(ADRO)can identify the disturbance change containing the back EMF information in real time,and then accurately estimate the motor speed according to the observed disturbance.The ADRO can deal with the disturbances such as unmodeled dynamics,model parameter uncertainty and external disturbance.In this way,the system uncertainty and external disturbance can be estimated online so that the ADRO is robust to external disturbances and model errors.However,the parameter tuning of the ADRO is difficult,and the fixed parameters cannot realize the optimal control under different working conditions.Therefore,the fuzzy control and ADRO are combined to realize the adaptive parameter tuning of the ADRO in this dissertation,and a novel FADRO for the PMSM sensorless control is proposed to combine the characteristics of adaptive parameter adjustment of the fuzzy control and disturbance compensation of the ADRO.The FADRO can accurately identify the back EMF and eliminate the influence of parameter tuning and system uncertainty disturbance.The experimental results show that the proposed sensorless control system can obtain good speed estimation and control performance,and possesses strong robustness to system parameter mismatch and external disturbance.Finally,in order to suppress the current harmonic and noise of the PMSM,a harmonic suppression strategy based on the improved resonant controller is proposed to further optimize the speed control,current control and sensorless control of the PMSM.The resonant current control optimized by the linear quadratic regulator(LQR)is used to separate the harmonic voltage from the dc voltage.The harmonic compensation voltage can suppress the current harmonic,and the dc voltage as well as the sampling current without harmonic component can estimate the speed so as to reduce the estimated speed harmonic.Then,the stability of the improved resonant controller is analyzed from the perspective of sensitivity function,and the corresponding parameter tuning strategy is proposed to realize the stable control with good current harmonic suppression.The experimental results show that the proposed scheme has superior current harmonic and noise suppression effect,and improves the system efficiency to a certain extent.