| Permanent magnet synchronous motor(PMSM) has the advantages of simple structure, reliable operation, in all aspects of the national economy, daily life, military industry, aviation and spaceflight, it has been widely used. The rotor position and speed are required for the implementation of the PMSM drive system. In high performance applications, the rotor position can be measured by the position sensor, but the installation of mechanical sensors will increase th e cost of the whole control system, and may reduce the system reliability. Therefore, the research of PMSM sensorless control system has important research significance and broad application prospect.The method of estimating the rotor position based on th e sliding mode observer is easy to implement in engineering, it has good disturbance rejection and simple control algorithm. This paper analyzes the principle of cla ssical sliding mode observer, in order to decrease the buffeting problem existing in the permanent magnet synchronous motor, we use the saturation function and the hyperbolic tangent function to replace the switch function method for inhibiting buffeting.Due to fever and other reasons, the stator parameters may change in the running process of the motor, this will affect the performance of the motor, this paper use the stator resistance online identification and parameter compensation to improve the robustness of the control system.In order to verify the effectiveness of the improved method, th e simulation model of PMSM sensorless control system based on Matlab/Simulink simulation platform is established. Classical sliding mode observer and improved sliding mode observer is simulated and analyzed respectively, the feasibility and superiority of the sensorless control strategy for the PMSM with the improved sliding mode observer are proved theoretically. From the simulation results, we find that the errors of the rotor position and speed are smaller with the improved sliding mode observer, it is more accurate. In the end, the experimental platform based on the XC164 CM control chip is carried out to verify the correctness and feasibility of the improved sliding mode observer. |