Model predictive control for a modular multilevel cascade converter with an improved zero-sequence voltage injection method

Abstract

In this study, the authors propose a model predictive control method for modular multilevel cascade converter-single star bridge cells (MMCC-SSBC) with an improved zero-sequence voltage injection algorithm under unbalanced conditions. First, a model predictive controller is designed to optimise the control process, and a novelty cost function is proposed to control the current, voltage and switching frequency. Second, an improved dc voltage balancing algorithm based on zero-sequence voltage injection is proposed. New unknown variables are selected to avoid the inverse trigonometric calculation and convert the non-linear problem to a linear problem. Third, the regulation capability is analysed to avoid the saturation and non-linear characteristic caused by the zero-sequence voltage injection. Simulation results indicate that the proposed algorithm shows a shorter calculation time and achieves better performance when single-phase voltage sag occurs. The experimental results further demonstrate that MMCC-SSBC combined with the proposed controller has excellent steady state and dynamic performance; at the same time, the switching frequency can be controlled by the new cost function.