The Study Of PMSM Control Methods Without Speed Sensor

Rotor position and speed are two essential parameters in the PMSM speed control system. The accurate measurements can be obtained by sensor installed on the shaft, but the sensor will increase system cost, increase the motor size and reduce system reliability. The control system without speed sensor can estimate the rotor position and speed through a certain method,and it requires only current, voltage and other physical quantities. By eliminating the speed sensor, the speed control system without sensor has simple structure, small size, low cost and high reliability. In this paper, the research object is PMSM with the surface type, and model reference adaptive and sliding model variable method is used to obtain the rotor position and speed. When load appear disturbance, motor speed will appear fluctuations. So the speed compensation and fuzzy PI are used to improve system performance. This thesis mainly does the following job:1. After the study of vector control theory and the principle of SVPWM, the PMSM vector control system is constructed.2. In order to inhibit the speed fluctuation, the load observer is constructed and used as the basis of speed compensation. The simulation results show that this method can effectively suppress speed fluctuation.3. After probing deeply into the theory of model reference adaptive theory, motor speed and rotor position are estimated through the model reference adaptive method. Control system simulation model is constructed based on this method, and the structural simulation results verify the validity of this method.4. After probing deeply into the sliding model variable structure control system, sliding-mode observer is used to obtain the rotor position and speed. Control system simulation model is constructed based on this method. The simulation results verify the validity and rationality of this method. 5. A kind of fuzzy PI controller auto-tuning is designed according to the fuzzy theory, and be used to replace the conventional PI controller on the speed loop. After comparing the two controllers, the system using fuzzuy PI shows better performance.