Improved Finite-Control-Set Model Predictive Control of a Modular Multilevel Converter with Reduced Computational Burden

Progress in high-power voltage source converter applications is driven by the Modular Multilevel Converter (MMC) and its derivatives. The unique characteristics of Model Predictive Control (MPC) have drawn the attention of power electronics researchers to its application in power electronic inverters. Implementing MPC in MMCs with numerous Sub-Modules (SMs) per arm poses challenges. This study introduces an Improved Folding MPC (IFMPC) strategy, coupled with a pre-processing sorting algorithm, to alleviate the computational burden. The proposed approach optimizes AC, DC, and circulating current simultaneously in a unified cost function. Real Time Hardware-in-the-Loop (HIL) results demonstrate satisfactory MMC response across diverse conditions with the proposed technique. The application of MMC, even with a high number of SMs per arm, becomes viable due to a substantial reduction in switching states and computational burden using the proposed MPC method.