The Research And Simulation Of Permanent Magnet Synchronous Motor's Control Strategy

Comparing with other motors, permanent magnet synchronous motor (PMSM) has its own excellent control properties, and has been applied to social activities popularly. What's more, there are quite abundant natural resources about the rare earth of permanent magnet in our country, which makes the research of PMSM's control very important. This article studies the control strategy of PMSM, which offers enough theoretical basis to practical application.The article analyzes the structure and characteristic of PMSM, and introduces three common coordinate series and the relationship of their mutual transformation in the PMSM's control. Moreover, it provides mathematical models of PMSM under the different coordinate series, especially introduces the mathematical model under d-q coordinate series.On this basis, it separately studies the control strategy of PMSM under speed sensor and under no speed sensor. Under the control of speed sensor, it studies the principles and achieving methods of three control strategy of space vector control(SVPWM), direct torque control(DTC) and direct torque control based on the modulation of space vector(SVM-DTC). Under Matlab/simulink, the simulation models of system are separately constructed and simulated. Under no speed sensor, it studies the principles and achieving methods of direct torque control strategy based on the speed identification of magnetic flux linkage vector angle and vector control strategy based on the speed identification of MRAS. Meanwhile, the simulation models of system are separately constructed under Mstlab/simulink and simulate to verify the theory.Through above-mentioned studies, it affirms that there are something deficient in various control strategies. In view of these deficiencies, it puts up further study and proposes some measures of optimization and improvement. Since the pulse of the torque and magnetic flux linkage are biggish in direct torque control, it puts forward some improvements such as optimizing the on-off table of voltage vector, improving the observer of magnetic flux linkage and so on. However , in no speed sensor control, the speed estimation has a big error when the torque changes, so it puts forward some improvements such as adjusting by changeable parameters, adding the compensation of speed and so on. Meanwhile, those improvements are verified in Matlab/simulink, and the results show its validity and availability.