Study On The Sensorless PMSM Vector Control System

In conventional motion control systems, optical encoders, hall sensors or eddy current sensors are used for this purpose. However, these additional sensors increase the cost and cubage and decrease the reliability. In this background, sensorless operation becomes hotspot in the control study of PMSM.Motor basic equation method and Model Reference Adaptive method of estimating the rotor speed and position are simulated by using MATLAB/SIMULINK. The former method has better dynamic performance, but the latter method has better steady performance. So the mix method is put forward in this paper. It combines motor basic equation method and Model Reference Adaptive method. The simulation results prove that the mix method has good dynamic and steady performance at the same time.An amended method of estimating the rotor speed about Model Reference Adaptive System(MRAS) on the basis of vector control is put forward. It only needs the error signal between q axis estimated current and q axis actual current. Then estimated speed is gained by using PI regulator which adjusted error signal. The experiment results prove that it possesses good performance of static and dynamic. Besides the other method of estimating the rotor speed based on Model Reference Adaptive System for PMSM is presented. Current model of PMSM is regarded as reference model while estimated stator flux model is considered as adjustable model. The adaptive laws are designed so that rotor speed and stator resistance are estimated at the same time. The experiment results prove that it is validity and feasibility.