Research On PMSM Sensorless Control System Based On Super-Twisting Sliding Mode Observer

In comparison with other motors,PMSM has many advantages,such as compact structure,reliable performance,high efficiency,high power density,low loss and flexible shape design,which make it more and more widely used.Recently,in order to reduce the cost of speed control system,to improve the reliability and to extend PMSM application field,sensorless control has become a research hotspot gradually.However,the velocity,accuracy and stability of rotor estimation influence the performance of the sensorless control system directly.Therefore,in this paper,the main research content is about these three aspects of sensorless control system of permanent magnet synchronous motor.Firstly,the rotor structure of permanent magnet synchronous motor is introduced briefly,and the surface permanent magnet synchronous motor is selected as the object in this paper.The mathematic model is analyzed and established in two-phase static coordinate system.On this basis,a simulation model of vector control system of surface permanent magnet synchronous motor in MATLAB/Simulink was built.The performance of the vector control system was verified by simulation.The simulation verifies that the vector control system has good dynamic performance of the motor speed adjustment,enabling the motor speed to track the given speed in real time.Then,the estimation principle of rotor position and speed based on back-EMF was elucidated.A second order Super-Twisting sliding mode observer(STSMO)is proposed to estimate the back-EMF of permanent magnet synchronous motor,Because the traditional first order sliding mode observer buffeting is relative seriously.In this paper,a second-order SuperTwisting sliding mode observer(Super-Twisting sliding mode observer,STSMO)was proposed to estimate the back electromotive force of permanent magnet synchronous motor,which can suppress the system chattering obviously.Besides,in order to solve the inherit chattering,an improved constant speed approaching law was adopted.At the same time,the stability of the commonly used conventional orthogonal PLL at present will change with the change of motor steering,resulting in a sharp increase in rotor position estimation error and poor system stability.Therefore,a phase locked loop based on arctangent function is proposed in this paper.The stability of the phase locked loop can be consistent with the motor forward and reverse,and the motor speed can be estimated accurately.Finally,the good dynamic performance of permanent magnet synchronous motor sensorless control system based on the improved Super-Twisting sliding mode observer was verified by the simulation.Finally,in the respect of the harmonic of the surface permanent magnet synchronous motor(SPMSM)stators’s back electromotive force will decrease the accuracy of rotor position estimation,so the harmonic component of the stators’ back electromotive force is analyzed in this paper firstly.And then,a second order generalized integrator is designed to filter the high frequency harmonics and dc components.The results of Fourier analysis of the filtered inverse electromotive force shows that the filter can suppress the high frequency harmonics and filter the dc components effectively,and improve the accuracy of rotor position estimation to a certain extent.