The Research And Optimization Of Sensor-less Control Algorithm For PMSM

In high-performance permanent magnet synchronous motor control system, it generally needs to install mechanical sensors on the rotor shaft, measuring the speed and position of the motor. But due to the influence of installation, environment, working time and other factors, there are many problems in the applications of mechanical sensors, so there is need to research on the sensor-less control of permanent-magnet synchronous motor.In this regard, this paper has established an integrated simulation platform for sensor-less control of permanent magnet synchronous motor, done a comparative study of the variety of sensor-less control algorithm, and improved the unscented Kalman filter algorithm. Then this paper verified the research with the real-time simulation platform dSPACE. Expected this research will play a positive role in practical engineering application. The specific research contents are as follows:1) research on the mathematical model of permanent magnet synchronous motor and vector control strategy, then establish the Simulink simulation model of permanent magnet synchronous motor;2) take the Inner-mounted permanent magnet synchronous motor(IPMSM) as the control object, research on four kinds of speed sensorless control strategies: model reference adaptive system(MRAS), rotating high frequency signal injection method,extended Kalman filter(EKF) and unscented Kalman filter(UKF); establish the corresponding control system simulation based on Simulink model, and compare the control effect of these control strategies;3) in respond to reduce the divergence risk existing in the calculation process of the unscented Kalman filter algorithm, research on the improved algorithm-square root unscented Kalman filtering algorithm, and further refine it to improve the tracking response capability and the stability of the algorithm. Establish the Simulink simulation model of improved speed sensorless control algorithm for permanent magnet synchronous motor, analyze and compare its control effect;4) simulate the actual permanent magnet synchronous motor in use of the hardware in the loop simulation function of dSPACE, verify the control effect of the speed sensorless control strategies mentioned above.Simulink simulation and dSPACE experimental results show that the improved square root unscented Kalman filtering algorithm can estimate the state of permanent magnet synchronous motor with high precision, response to the change of state or the external disturbance more quickly, and improve the stability compared with the original algorithm. The sensorless control system of permanent magnet synchronous motor based on this improved algorithm has good static and dynamic performance, and its feasibility has been verified in practice.