Controle preditivo baseado em modelo com abordagem da teoria de grafos para aplicação em conversores multiníveis CHB-B2B

Abstract

This work investigates the application of model-based predictive control (MPC) in conjunction with graph theory to address the inherent short-circuit states of the multilevel back-to-back cascaded H-bridge converter (CHB-B2B). Compared to classical back-to-back multilevel topologies, CHB-B2B offers a significant reduction in components per output voltage levels. However, this benefit is negated by the occurrence of internal short circuits during conventional PWM operation. The proposed approach leverages graph theory to define a CHB-B2B converter switching matrix, enabling the utilization of an appropriate control strategy, such as MPC. This strategy not only eliminates the detrimental short-circuit stages but also explores all remaining valid switching states, thereby preserving converter controllability and power quality. The effectiveness of the proposed MPC-graph theory approach is validated through real-time experimental results obtained on a hardware-in-the-loop (HIL) platform. The experiments encompass four distinct solid-state transformer (SST) configurations using CHB-B2B structure: input-parallel output-parallel (IPOP), input-parallel output-series (IPOS), input-series output-parallel (ISOP), and input-series output-series (ISOS). These configurations cater to distribution grids with diverse voltage and power levels. The results convincingly demonstrate the strategy’s efficacy in regulating DC-link voltages, synthesizing multilevel output voltages, and achieving currents with minimized harmonic distortion