A Fast Finite-Level-State Model Predictive Control Strategy for Sensorless Modular Multilevel Converter

Abstract

In this article, we propose a fast finite-level-state model predictive control (FFLS-MPC) strategy for solving the well-known challenges in the predictive control-regulated modular multilevel converter (MMC), which aims to overcome the high computational complexity and enhance the reliability of the control system. First, the presentable output voltage level is determined from solving the Diophantine equations' point of view, which has a good potential to avoid the online optimization process and the selection of weighting factors compared with the conventional finite control-set model predictive control (FCS-MPC) approach. In this sense, the overall computational effort is significantly decreased while achieving satisfactory control performance. Alternatively, in order to enhance the system reliability under sensor failures (e.g., arm current sensor and submodule (SM) voltage sensor), a novel control strategy for sensorless MMC is presented by combining an adaptive linear-neuron-based SM voltage estimation scheme with a currentless sorting-based capacitor-voltage-balancing approach to serve this purpose. We contribute two main points to the relevant literature. The first one is the improvement of computationally efficient by employing the proposed FFLS-MPC methodology. The second one is the elimination of the arm current sensor and SM voltage sensor while guaranteeing adaptability to different conditions. Finally, compared with the state-of-the-art FCS-MPC strategies, comprehensive simulation studies and experiments are presented to demonstrate the effectiveness and feasibility of the proposed methodology for MMC.