Model-Predictive Flux Control of 3L-NPC Fed Induction Motor Drives with Stator Flux Error Based Two-stage Optimization

This study proposes a simplified flux-error based voltage vector selection method for predictive flux control of a three-level neutral point clamped inverter (3L-NPC) fed induction motor. The controller of a 3L-NPC fed IM drive has 27 voltage vectors for cost function evaluation. The proposed optimization method reduces the number of voltage vectors from 27 to 14 for prediction and minimization of cost function by using a two-stage optimization technique. The reference stator flux vector calculation (RSFVC) technique is employed in order to reduce the computational complexity of the prediction further, by regulating the torque outside of the cost function. Compared with conventional finite state predictive torque control (FS-PTC) for 3L-NPC VSI fed induction motor drive, the proposed method reduces the computational time by 40%. The dynamic and steady-state performances of the proposed method are investigated and compared with the conventional FS-PTC in terms of electromagnetic torque and stator flux responses, total harmonic distortion of stator currents, neutral point voltage variations and robustness of the drive.