湖北师范大学学报(自然科学版)

为解决有限控制集模型预测控制(FCS-MPC)计算量大和权重因子调整繁琐的问题，提出了一种无需权重因子的快速模型预测控制方法。该方法利用无差拍思想构造期望参考电压矢量，并选择中性点钳位型逆变器(NPC)作为拓扑模型。通过参考电压矢量的相位角确定扇区，仅需枚举扇区的4个基本电压矢量代替传统的27个，显著减少了计算量，同时选择合适的冗余小矢量平衡中性点电压。该算法简化了控制目标，无需调整权重因子，提升了算法的可移植性。MATLAB/SIMULINK仿真验证了该方法的可行性和有效性。

To address the issues of high computational burden and the complex tuning of weighting factors in Finite Control Set Model Predictive Control(FCS-MPC), this paper proposes a fast model predictive control method without weighting factors. This method constructs the desired reference voltage vector based on the deadbeat control concept and adopts the Neutral Point Clamped(NPC) inverter as the topology model. It uses the phase angle of the reference voltage vector to determine the sector, and enumerates only four basic voltage vectors within the corresponding sector instead of the conventional 27 vectors, which significantly reduces the computational load. Additionally, it selects appropriate redundant small vectors to balance the neutral point voltage. The proposed algorithm simplifies the control objectives, eliminates the need for weighting factor tuning, and enhances portability. It also verifies the feasibility and effectiveness of the method through MATLAB/SIMULINK simulations.