Investigation Of Sensorless Vector Control System For PMSM Based On High Frequency Signal Injection

Permanent magnet synchronous motor (PMSM) has been widely used in a variety of high performance drive systems for its advantages, such as high efficiency, high power density and runnability, etc. To achieve the high performance control requirements of PMSM, vector control is used usually, which requires accurate rotor position to realize field orientation. The traditional methods are through various sensors to get the rotor position, however, the sensor will not only increase the system cost, reduce the system reliability, but also limit the application of PMSM in some cases. In the thesis, the rotating high-frequency voltage signal injection method is selected to achieve PMSM sensorless vector control, which is on the basis of the classification and comparison of the existing sensorless control technology.Firstly, the structure and mathematical model of PMSM are analyzed, based on which the principle of PMSM vector control is described in detail. According to the characteristics of the control object, the two current control strategies, id=0 and maximum torque per ampere (MTPA), are contrasted and analyzed.Secondly, the principle of the high-frequency voltage signal injection method to estimate the rotor position is described, which focuses on the high-frequency carrier current signal demodulation, using PLL technology to achieve the estimate of rotor position, determining the magnetic polarity of the rotor with the saturation effect and multiple salient effects of rotor position estimation.Thirdly, the permanent magnet synchronous motor sensorless vector control system was designed. Piecewise curve fitting method is used to realize the MTPA control strategy approximately. The shortcomings of conventional PI regulator is analyzed, combining with fuzzy control, a fuzzy adaptive PI regulator is designed to replace the traditional PI speed controller to optimize the system control.Finally, the simulation model of PMSM sensorless vector control system is established by using the software of MATLAB/Simulink. Simulation experiment is operated on the simulation model; the results show that by using high frequency voltage signal injection method the rotor position can be accurately estimated in medium and low speed, especially in low speed, the effect is better. At the same time, to contrast with conventional PI control, the system has better anti-disturbance due to the fuzzy adaptive PI speed controller, and the whole control system has good static and dynamic performance.