Parameters Identification of IPMSM Based on Deadbeat Predictive Current Control

This paper proposes a parameters identification method for the interior permanent magnet synchronous motor (IPMSM). The sinusoidal harmonic are injected into torque control angle of IPMSM to observe the motor parameters. By extracting the fundamental frequency and high frequency signals of the stator voltage, the motor parameters are obtained by using the recursive least square (RLS) algorithm. Deadbeat predictive current control is used to track the harmonic command values of the currents. In this paper, the interference of harmonic currents to the observation of inductances is analyzed while the injection of harmonic currents will not affect the identification result of inductances. Moreover, the maximum torque per ampere (MTPA) control method is used to reduce the influence of injected harmonic on torque ripple. Finally, the feasibility of this method is verified by simulation.