Optimal Extended Phase-Shift Control with Residual Reinforcement Learning for DAB Converter Efficiency Maximization
Authors:
- Nebiyeleul AmareNANebiyeleul Amare
- Donghun LeeDLDonghun Lee
- Nguyen NamNNNguyen Nam
- Naehyuck ChangNCNaehyuck Chang
- Young-Il LeeYLYoung-Il Lee
Preprint
- 22 September 2026
- preprint
- Published by Elsevier in SSRN Electronic Journal
Abstract
Conventional extended phase-shift (EPS) control methods for dual-active-bridge (DAB) converters are commonly designed under fixed voltage-conversion conditions, which limits their ability to exploit voltage-ratio-dependent efficiency improvement. In addition, accurate closed-loop power regulation remains challenging due to nonlinear switching dynamics and the high-frequency alternating inductor current. This paper proposes an optimal EPS control framework with residual reinforcement learning (RL) adjustment, in which offline optimization jointly determines the EPS parameters and voltage conversion ratio, while residual RL provides online corrections to enhance output-power regulation. The joint optimization is formulated to improve efficiency by minimizing backflow power and converter losses subject to zero-voltage switching (ZVS) constraints. A Twin Delayed Deep Deterministic Policy Gradient (TD3)-based residual RL controller is applied on top of the precomputed optimization-based baseline to compensate for tracking errors and operating uncertainties. To improve operational safety, the RL action space is constrained by hard boundaries derived from the steady-stateoptimization results. Experiments on a DSP-controlled DAB hardware platform demonstrate that the proposed method improves tracking performance compared with optimal baseline EPS, PI, and MPC methods, maintains stable operation under parameter variations, maintains ZVS operation, and achieves higher efficiency improvement over related methods without joint optimization.Keywords
This publication has 22 references indexed in Scilit:
- Near real-time online reinforcement learning with synchronous or asynchronous updatesScientific Reports, 2025
- Deep reinforcement learning-based proportional–integral control for dual-active-bridge converterNeural Computing and Applications, 2023
- A Novel Intelligent Nonlinear Controller for Dual Active Bridge Converter With Constant Power LoadsIEEE Transactions on Industrial Electronics, 2022
- Minimum Backflow Power and ZVS Design for Dual-Active-Bridge DC–DC ConvertersIEEE Transactions on Industrial Electronics, 2022
- Deep Reinforcement Learning-Aided Efficiency Optimized Dual Active Bridge Converter for the Distributed Generation SystemIEEE Transactions on Energy Conversion, 2021
- Model Predictive Control of Power Electronic Systems: Methods, Results, and ChallengesIEEE Open Journal of Industry Applications, 2020
- Minimum-Backflow-Power Scheme of DAB-Based Solid-State Transformer With Extended-Phase-Shift ControlIEEE Transactions on Industry Applications, 2018
- Next-Generation Multi-Functional Modular Intelligent UPS System for Smart GridIEEE Transactions on Industrial Electronics, 2012
- Accurate Power Loss Model Derivation of a High-Current Dual Active Bridge Converter for an Automotive ApplicationIEEE Transactions on Industrial Electronics, 2009
- Practical PID Control2006