Robust Sensorless Direct Speed Predictive Control of Synchronous Reluctance Motor

To achieve a fast dynamic response, sufficient load response, and good tracking performance, a precision and robust method for sensorless direct speed predictive control (DSPC) of synchronous reluctance motor (SynRM) is proposed and simulated in this paper. The proposed method replaces the cascaded control structure (i.e. speed and currents loops) of the field-oriented control to a strategy that exploits all the electrical and mechanical variables in one control law to select the optimal switching vector for the two-level inverter to apply in the next sampling interval. Furthermore, for robustness, an extended Kalman filter (EKF) is used to observe rotor speed and position, stator currents, and mechanical load torque, and consequently, the system reliability is improved and the cost is decreased. Then, the observed variables are fed back into the prediction model. For simplicity, particle swarm optimization (PSO) is applied to tune the weighting factors of the cost function and EKF covariance matrices. Simulation results reveal the robustness and reliability of the proposed method.