Long-Horizon Robust Direct Model Predictive Control for Medium-Voltage Drives with Active Neutral-Point Potential Balancing

The paper presents a direct model predictive control algorithm for medium-voltage (MV) induction machines driven by three-level neutral-point-clamped (NPC) inverters that incorporates the neutral point potential balancing. For such nonlinear systems implementation of long horizons may be regarded as even a formidable task due to the high computational complexity. Nevertheless, this can be achieved with a modest calculation cost by decreasing the size of the underlying control optimization problem. Moreover, when assisted by a light estimation algorithm, the developed control scheme achieves a high level of robustness to variations in the motor parameters. The presented results demonstrate the effectiveness of the proposed method during steady-state and transient operating conditions.