Answers are grounded in this paper and cite their sources.
Reinforcement learning improves voltage regulation in LLC resonant converters, indicating enhanced performance under various conditions.
To address the limited dynamic voltage regulation performance of LLC resonant converters under wide input voltage and load variations, a reinforcement learning-based voltage control strategy is proposed in this paper. The twin delayed deep deterministic policy gradient (TD3) algorithm is adopted to learn the nonlinear mapping between system states and control actions, enabling adaptive adjustment of the converter operating parameters. Based on the established LLC resonant converter simulation model, the state space, action space, and reward function of the agent are designed to ensure rapid control response to abrupt changes in input voltage and load. Compared with the conventional PI control strategy, the proposed TD3-based strategy provides faster control actions during operating condition transitions, effectively suppressing output voltage overshoot and undershoot, and shortening the settling time. Simulation results verify that the proposed method achieves improved dynamic response performance under various operating conditions, demonstrating its effectiveness and superiority in LLC resonant converter voltage regulation.
No takes yet. Share an insight, caveat, or question.
Pan et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: