Advanced Model Predictive Control Strategy in Optimization of Permanent Magnet Synchronous Machine

In this paper, an optimized model predictive control (MPC) strategy for a permanent magnetic synchronous machine (PMSM) system is introduced and a PMSM fed by a three-phase two-level voltage sourced inverter is used to introduce two different MPC strategies: direct predictive control (DPC) and pulse width modulation predictive control (PPC). In the proposed MPC strategy, the model with sampling time delay and its compensation method is taken into consideration. Furthermore, Kalman Filter is introduced to observe the disturbances due to the mismatch of the real machine parameters and the measured values. In the last scenario, switching losses as well as the THD results are both considered so as to find an optimal frequency reducing both the value of THD and the switching losses. Performance of the proposed control strategy is verified using MATLAB/Simulink. The simulation results confirm the effectiveness of the proposed MPC strategy in confining the total harmonic distortion value of current of the system and in tracking the optimal referenced torque from the PMSM.