An Improved Duty Cycle Modulation‐Based Novel Reverse Model Predictive Voltage Control Strategy for Modular Multilevel Converter

ABSTRACT The conventional model predictive voltage control strategy for modular multilevel converters (MMCs) suffers from the difficulty of weighting‐factor design and the high computational burden caused by switching‐state evaluation and capacitor‐voltage sorting. To address these issues, an improved duty‐cycle‐modulation‐based reverse model predictive voltage control (IDC‐RMPVC) strategy is proposed in this paper. First, the AC‐side output current and circulating current are selected as the main control objectives. Based on the discrete‐time predictive model of the MMC, the required arm‐voltage difference and arm‐voltage sum are directly obtained through inverse prediction. Then, the corresponding upper and lower arm duty cycles are calculated and implemented by carrier phase‐shifted modulation, avoiding finite switching‐state enumeration and multiobjective cost‐function optimization. In addition, a bounded duty‐cycle regulation signal is injected into the modulation wave to achieve submodule capacitor‐voltage balancing without capacitor‐voltage sorting. The practical implementation also considers dead‐band regulation, arm‐current zero‐crossing treatment, and duty‐cycle saturation. Compared with the R‐MPVC strategy, the proposed IDC‐RMPVC reduces the capacitor voltage ripple by 12.8%, the output voltage THD by 36.8%, and the output current THD by 22.8%. Experimental results based on the RT‐LAB hardware‐in‐the‐loop platform verify the effectiveness of the proposed strategy.