Research On Position Sensorless Of Permanent Magnet Synchronous Motor

Permanent magnet synchronous motor (PMSM) owing to its smaller size, simplestructure, smaller loss, smaller moment of inertia, reliable performance, strong ability to adaptto the environment, easy control and maintenance, etc., have gradually widely used in ACdrive system. The position sensorless PMSM control system can not only reduce costs, butalso improve the reliability of the system. Thus, in recent years, position sensorless control ofPMSM become a hot research field, many different control strategies and algorithms havebeen proposed at home and abroad. For position sensorless technology can not realizeobservation of the whole speed range at present stage, it studies the hybrid control methods indifferent speed range of position sensorless technology, achieving the full speed rangesensorless control in PMSM drive system, the main researches of the thesis is on thefollowing areas:(1) After the inspection of the relevant literature, the description of the advantages ofposition sensorless control of PMSM, classifies and summarizes the methods for estimation ofthe current position sensorless PMSM. Finally, and the proposition of the composite methodof the estimation for position and speed at full speed range for PMSM.(2) The thesis introduces the classification and structure of the PMSM, and establishesmathematical model of a three-phase PMSM in the stationary coordinate system ABC,two-phase stationary coordinate system αβ, two-phase rotating coordinate system dq. Itdescribes the PMSM control strategy. Based on the two-phase rotating coordinate system dq,the simple id=0control strategy of vector control system is selected, which lays the foundationfor the subsequent observer study.(3) A detailed description and analysis of the model reference adaptive system (MRAS)observer for PMSM position and speed estimation method is introduced. For the MRAS relieson accurate motor model, the motor’s sensitive characteristic to parameters and disturbances,variable structure theory is proposed into MRAS adaptive rate, variable structure MRASposition and speed observer is designed. On this basis, the model of PMSM positionsensorless vector control system is built and simulated in MATLAB/Simulink.(4) For MRAS observer is not good at estimation of PMSM position and speed where inits low speed, a composite estimation method of position and speed is proposed, which isbased on pulsating high-frequency voltage signal injection method and MRAS. Estimation ofposition and speed at low speed uses pulsating high-frequency voltage signal injection methodand at high speed uses MRAS, and ultimately achieves accurate estimation and control withinthe full scope of the position and speed. MATLAB/Simulink simulation of hybrid position and speed identification method can run in full speed range.