面向效率提升的双有源桥DC-DC变换器六自由度调制方案及深度强化学习优化

The conversion efficiency of a dual-active-bridge (DAB) converter could be improved by increasing the number of control degrees of freedom (DoFs). A modulation with six-degree-of-freedom (6-DoF) is proposed in this article to further improve the efficiency of DAB converters. First, the frequency-domain analysis is used to directly derive the operating expressions of the converter, including the transmission power, inductor current, and root-mean-square (rms) current. Then, a deep reinforcement learning (DRL) optimization scheme with 6-DoF (DRL-6DoF) is proposed. Specifically, the deep deterministic policy gradient (DDPG) algorithm is used for solving optimal solutions with the minimum power losses. The trained DDPG agent can output the optimal values of control variables under various operation conditions. With the proposed DRL-6DoF control scheme, the efficiency of the DAB converter is further improved because of the low rms current and great zero-voltage switching (ZVS) performance. Especially under heavy load conditions, the efficiency is greatly improved because of the characteristic of variable switching frequency. Finally, the effectiveness of the proposed method is verified by experimental results on a built prototype.