Research On Permanent Magnet Synchronous Motor Vector Control System Based On Dead-Time Compensation

Permanent magnet synchronous motor (PMSM) has received extensive attention for its unique structure and excellent performance, which adapts for more and more high-performance applications. This article is mainly focusing on improving the performance of PMSM control system. Vector control of PMSM, space vector pulse-width modulation (SVPWM) and inverter dead-time compensation are studied deeply respectively.Based on the mathematics model of PMSM, the theory and realization of vector control have been clarified, and vector control system of PMSM is established. As the key of vector control system, a detailed study of SVPWM technology is carried out. This method can effectively improve the current waveform, reduce motor torque ripple, and be easy to digital realization. Vector control system of PMSM is simulated by MATLAB/Simulink, which is shown good steady and transient performance.Also, in order to enhance the capability of vector control system of PMSM, an in-depth study of inverter dead-time compensation is carried out. Based on extensively analyzing and summary of the existing research on dead-time compensation home and overseas, three new practical methods of dead-time compensation are proposed: phase compensation, the average voltage of current-feedback compensation, coordinate transformation compensation. At the same time, forecast-current method and current-vector method are presented to solve the problem of current polarity detection such as zero-crossing-point fuzzy, low precision and implementation complexity, and so on in the past. The simulation results with MATLAB/Simulink prove that the methods above can improve current and output electromagnetic torque waveform of the motor stator evidently.In order to further validate the study of vector control system of PMSM and dead-time compensation method, simulating experiments are accomplished by the hardware-in-loop (HIL) simulation platform of dSPACE. Firstly, the component and function of the HIL platform are described concretely. Secondly, the real-time models of HIL simulation of the system are established. At last, the major steps of experiment of HIL simulation are presented, and experimental waveforms are recorded by the oscilloscope, which prove that dead-time compensation methods presented in this paper are correct, feasible and effective.