Artificial Intelligence-Aided Optimization Strategy for Three-Level DAB Converters Modulated with Five DoFs

The three-level dual-active-bridge (DAB) converter presents several benefits, including enhanced control flexibility and the ability to use power switches with lower voltage ratings compared to traditional two-level DAB converters. To improve transmission efficiency, an optimization approach based on deep reinforcement learning (DRL) has been proposed, leveraging the five degrees of freedom (5- DoFs) control method. This approach applies the deep deterministic policy gradient (DDPG) algorithm to minimize power losses and determine optimal control strategies. The trained DDPG agent functions as a predictor, facilitating appropriate control actions under varying operational conditions. The main goal is to maximize efficiency by minimizing power losses. With the proposed strategy, the conversion efficiency is well improved, and the effectiveness is verified by experiment results under various conditions.