High-Precision Sensorless Predictive Control of Salient-Pole Permanent Magnet Synchronous Motor based-on Extended Kalman Filter

This paper presents an accurate method for sensorless Current Predictive Control (CPC) of a salient-pole permanent magnet synchronous motor (PMSM) supplied from two-level power converter. For the sensorless process, Extended Kalman Filter (EKF) in the rotating reference frame is used as a non-linear position, speed and currents observer for PMSM. For uncomplicatedness and straightforwardness, Particle Swarm Optimization (PSO) is used instead of conventional trial and error method to obtain the accurate parameters of the PI speed controller and the EKF Covariance matrices. In addition, Current predictive control is used instead of the PI current controllers in the conventional Field Oriented Control (FOC). CPC uses a discrete model of PMSM to predict the currents in the future for all the possible switching vectors and with the aid of cost function, the optimal switching vector is selected to apply in the next sampling interval. The proposed methodology is validated through MATLAB/Simulink to clarify the ability of the proposed control algorithm in different speed area including zero speed. Results demonstrate high performance and robustness of the drive with minimum total harmonic distortion.