Double-Vector Model Predictive Control for Single-Phase Five-Level Actively Clamped Converters

Abstract

In this article, a double-vector-based model predictive control (MPC) method with efficient computation for single-phase five-level active neutral-point clamped (5L-ANPC) converters is proposed. It uses double voltage vectors per control cycle rather than the single in the conventional MPC method, resulting in reduced output current ripples and fixed switching frequency. First, the MPC method based on the Lyapunov function is optimized to select two voltage sectors rather than 25, reducing the calculation complexity significantly. Then, two current tracking algorithms based on double voltage vectors are presented, which are using the current cost function and current slope, respectively. Finally, a fast voltage balancing scheme for the dc-link and flying capacitors is developed and embedded in the MPC method. Both simulated and experimental results are used to validate the correctness and feasibility of the proposed MPC method.