This article presents an optimal model predictive flux control (MPFC) for a two-level inverter fed induction motor. Integrating discrete SVM into FCS-MPFC enhances the performance of the IM drive. However, conventional DSVM-MPFC requires to enumerate and evaluate a higher number of virtual vectors in the prediction loop. In this paper, a high-efficient and low complexity voltage selection method is proposed to reduce the number of candidate voltage vectors from 38 to 15 without any suboptimality. Both steady-state and transient performances of the proposed method remain the same as the 38-vector based conventional DSVM-MPFC, producing the same cost-function values in all operating conditions. Furthermore, an online switching frequency reduction technique is proposed to achieve a minimum commutation per inverter vector change within each sampling cycle and between adjacent cycles. By appropriately arranging the sequence of real voltage vectors in each sampling cycle, a lower average switching frequency is achieved. The proposed switching frequency reduction method decreases the switching losses without compromising the performance of DSVM-MPFC as only the applied sequences of the real voltage vectors are optimized. Experimental studies are conducted to verify the effectiveness of the proposed algorithm.

中文翻译：

---

基于离散空间矢量调制的IM驱动器降低开关频率的模型预测通量控制

本文提出了一种用于两级逆变器馈电感应电动机的最优模型预测磁通量控制（MPFC）。将离散的SVM集成到FCS-MPFC中可增强IM驱动器的性能。但是，传统的DSVM-MPFC需要枚举和评估预测循环中的大量虚拟向量。本文提出了一种高效，低复杂度的电压选择方法，将候选电压矢量的数量从38个减少到15个，而没有任何次优性。所提出方法的稳态和瞬态性能都与基于38矢量的常规DSVM-MPFC相同，在所有工作条件下均产生相同的成本函数值。此外，提出了一种在线开关频率降低技术，以在每个采样周期内以及相邻周期之间实现每个逆变器矢量变化的最小换向。通过在每个采​​样周期中适当地排列实际电压矢量的序列，可以获得较低的平均开关频率。由于仅优化了实际电压矢量的应用序列，因此所提出的开关频率降低方法可在不损害DSVM-MPFC性能的情况下减少开关损耗。进行实验研究以验证所提出算法的有效性。由于仅优化了实际电压矢量的应用序列，因此所提出的开关频率降低方法可在不损害DSVM-MPFC性能的情况下减少开关损耗。进行实验研究以验证所提出算法的有效性。由于仅优化了实际电压矢量的应用序列，因此所提出的开关频率降低方法可在不损害DSVM-MPFC性能的情况下减少开关损耗。进行实验研究以验证所提出算法的有效性。