Virtual Vector-Based FCS-MPC for NPC Three-Level Grid-Tied Inverter Without Weighting Factor of Neutral-Point Voltage Balancing

The neutral-point clamped (NPC) inverter is widely used in energy conversion, but the unbalance of neutral-point voltage (NPV) will lead to imprecise control and the decline of power quality. Conventional finite control set model predictive control (FCS-MPC) uses an extra NPV weighing factor to balance the NPV. But the weighting factor should tune frequently to match the best value. To cope with this problem, this paper proposes a virtual vector-based control strategy without the NPV weighting factor for NPC inverters. The paper analyses the relationship between neutral-point current and neutral-point voltage. An NPV vector is designed to inject a compensatory neutral-point current to balance the capacitor voltage. The duration time of each NPV vector is calculated from the real-time unbalance of the NPV. Because each NPV balancing virtual vector contains one NPV vector, the proposed method can balance the NPV without an extra NPV weighting factor in the cost function. Besides, a cost function without the NPV weighting factor is designed for the proposed method. Finally, experiments are carried out to verify the validity of the proposed method.