Improved Sensorless Control For Pulsating High Frequency Signal Injection Of Permanent Magnet Synchronous Motor

Permanent magnet synchronous motor(PMSM)is widely used in various fields because of its high performance,simple structure,small size and other advantages,and its common control mode is vector control.In PMSM vector control system,the speed and position of the rotor need sensors to obtain,but the use of sensors will increase the volume and cost of the system,and have high requirements for the application environment.Therefore,sensorless control technology has become a research hotspot at home and abroad.The sensorless control at medium and high speed is mainly based on the fundamental wave model of the motor,and the research on this aspect is relatively mature,while the sensorless control method at zero low speed is mainly the high frequency signal injection method.However,in the process of position information extraction by the high-frequency injection method,the phase lag will be generated by the filter,which will affect the accuracy of rotor position estimation and the dynamic characteristics of the system.To solve these problems,this paper studies the high frequency injection method and improves the PMSM vector control system based on high frequency pulsating signal injection,so as to obtain better control performance.This paper firstly introduces the structure and physical model of PMSM,and compares the surface permanent magnet synchronous motor(SPMSM)with the built-in permanent magnet synchronous motor(IPMSM).Then,the mathematical model of PMSM in natural coordinate system is established based on the idealized hypothesis,and the mathematical model of PMSM in stationary coordinate system and synchronous rotation coordinate system is derived based on the relevant knowledge of coordinate transformation.Finally,the mathematical model under high frequency excitation is obtained.Secondly,the pulsating high frequency signal injection method is studied theoretically based on the high frequency mathematical model,and the PMSM vector control system model based on pulsating high frequency signal injection is established.In the section of position observation,the common position observer is introduced and the advantages and disadvantages of phase-locked loop and Luenberger observer are compared.At the same time,the selection of the frequency and amplitude of the injected signal and the problem that the convergence point is not unique when detecting the initial position of the rotor are studied.Finally,the PMSM vector control system based on pulsating high frequency signal injection is improved.Firstly,the influence of filter phase lag is analyzed,aiming at the problem that the filter phase lag increases the rotor position estimation error,the phase compensation is carried out for the bandpass filter.At the same time,the estimation error after the separation of the low pass filter has a small number of high frequency components.A new modulation function is designed to make the estimated error of the modulated signal more different from the frequency of the high-frequency component,and the filtered error signal has less high-frequency component content,which effectively reduces the estimation error of the rotor position.In order to solve the problem that the phase lag of the filter affects the dynamic performance of the system,from the perspective of controller,model predictive control(MPC)is introduced into the rotating speed loop instead of PI control,and its rolling optimization and feedback correction characteristics can greatly improve the dynamic performance of the system.At the same time,the common MPC was improved,and a softening coefficient scheme was designed to further improve the control performance.Secondly,in order to obtain better anti-load disturbance capability,this paper adopts load disturbance compensation strategy,analyzes the problem of traditional Q-axis reference current feedforward compensation under model predictive control,and proposes a new compensation method.This method draws on the idea of MPC,takes the observed load torque as the reference trajectory,and lets the predicted electromagnetic torque track the reference trajectory.The relationship between the predicted electromagnetic torque and the voltage compensation of the q axis is found,and the reference voltage of the q axis is compensated,and a better compensation effect is obtained.