A Simplified Modulated Model Predictive Control for Modular Multilevel Converter

Abstract

The Modular Multilevel Converter (MMC) has proved to be a revolutionary topology for application in high voltage domain. Model Predictive Control (MPC) has garnered a lot of attention in the recent years for the control of multiple input multiple output (MIMO) systems like the MMC. Though highly advantageous in terms of simplicity of implementation and fast dynamic performance, the computational complexity of MPC increases as the number of submodules (SMs) increase. Moreover, issues like the design of weights in cost function and variable switching frequency hinder the use of MPC for the MMC. To take care of this problem, a unique Modulated Model Predictive Control (M <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sub> pC) strategy has been implemented for the MMC using a simple search algorithm. The proposed approach takes care of the above issues in addition to ensuring control of output current, circulating current and capacitor voltages. The proposed method ensures fixed switching frequency with reduced calculations. The performance of the method is examined using simulation studies for a 10 kV/0.7 MW three-phase MMC system connected to a passive load.