Adaptive Control of Wireless Power Transfer for Electric Vehicle Charging under Variable Coupling Conditions Using Proximal Policy Optimization
DOI:
https://doi.org/10.68050/JAMS.2026.448Keywords:
Wireless power transfer, Electric vehicle charging, Proximal Policy Optimization, Reinforcement learning, Adaptive control, Coupling coefficient, Power-transfer efficiency.Abstract
Wireless power transfer (WPT) offers a convenient and flexible solution for electric vehicle (EV) charging, but its performance is highly sensitive to variations in coil separation and magnetic coupling. This paper proposes a Proximal Policy Optimization (PPO) adaptive control strategy for resonant WPT EV charging under variable coupling conditions. A series-series compensated WPT system is modeled, and a PPO controller is developed to regulate the charging current and operating frequency according to the instantaneous coupling coefficient, transfer efficiency, battery voltage, and charging current. The proposed controller is trained and evaluated in MATLAB/Simulink under coil separation distances ranging from 5 to 20 cm. The PPO agent begins to converge at approximately 1200 training episodes and reaches a stable learning regime after around 1600 episodes. Compared with conventional fixed-frequency control, the proposed strategy improves the WPT transfer efficiency from 89.3% to 91.0% at a 5 cm coil separation and from 68.2% to 73.9% at 20 cm. The improvement becomes more pronounced as the coupling weakens. The dynamic-response evaluation gives a settling time of approximately 0.48 s, with a maximum deviation of 5.2% and a steady-state error below 2.8%. These results indicate that PPO-based adaptive control can improve both transfer efficiency and dynamic adaptability of wireless EV charging under variable coupling conditions. Overall, the results demonstrate that PPO-based adaptive control can effectively improve efficiency and dynamic regulation of wireless EV charging systems, particularly when variations in coil separation lead to unfavorable coupling conditions. The proposed approach provides a promising foundation for intelligent WPT controllers capable of maintaining reliable charging performance under practical operating variations.
References
Chen, F, H Hu, L Zhao, A Padilla, and J Hou. 2023. “A Linear Parameter-Varying Hammerstein Model for Dynamic Modeling of WPT Systems.” IEEE Transactions on Power Electronics 38(12): 16230–16244.
2. Feng, Y, F Chen, Y Sun, T Lin, F Meng, and H Hu. 2024. “Identification and Control of LCC-S WPT Systems Using a Linear Parameter Varying Model.” IEEE Transactions on Power Electronics 39(9): 11862–11873.
3. Hao, L, and Y Xu. 2024. “Reinforcement Learning of Threshold Policy for Dynamic Wireless Charging with Distributed Energy Resources.” In Proc. IEEE PES General Meeting (PESGM) Seattle, WA, USA, 2024, Seattle, WA, USA, 1–5.
4. Jamjuntr, P, and P Suanpang. 2024. “Adaptive Multi-Agent Reinforcement Learning for Electric Vehicle Charging Networks.” World Electric Vehicle Journal 15(10): 453–468.
5. Park, K, and I Moon. 2022. “Multi-Agent Deep Reinforcement Learning Approach for EV Charging Scheduling in a Smart Grid.” Applied Energy 328: 120111.
6. S. A. Benfadhel, A, A J M Cardoso, and M Trabelsi. 2025. “Intelligent Optimization and Real-Time Control of Wireless Power Transfer for Electric Vehicles.” IEEE Access 13: 45872–45886.
7. S. A. Sagar, A. Kashyap, M. A. Nasab, S. Padmanaban, M. Bertoluzzo, A. Kumar, and M. Al-Haddad. 2023. “A Comprehensive Review of the Recent Development of Wireless Power Transfer Technologies for Electric Vehicle Charging Systems.” IEEE Access 11: 83703–83751.
8. Xu, X, Y Jia, Y Xu, Z Xu, and C S Lai. 2024. “Multi-Agent Reinforcement Learning-Based Data-Driven Energy Management for Electric Vehicle Charging Systems.” IEEE Transactions on Smart Grid 15(2): 1324–1338.
9. Xue, Z, W Liu, C Liu, and K T Chau. 2025. “Critical Review of Wireless Charging Technologies for Electric Vehicles.” World Electric Vehicle Journal 16(2): 65.
10. Zhang, S, R Jia, H Pan, and Y Cao. 2023. “A Safe Reinforcement Learning-Based Charging Strategy for Electric Vehicles in Residential Microgrid.” Applied Energy 348: 121490.
11. Zhang, Z, W Liu, and H Sun. 2025. “Model-Free Safe Deep Reinforcement Learning for Grid-to-Vehicle Charging Management.” Engineering Applications of Artificial Intelligence 145: 112529.
12. Zhao, S, F Chen, C Tang, P Deng, and C Duan. 2024. “Modeling and Control of WPT Systems in the Presence of Load and Mutual Inductance Variations.” IEEE Transactions on Power Electronics 39(11): 15315–15328.
13. Zhao, Z, S Tang, C Chen, D Zhao, and P Deng. 2023. “Modeling and Control of the Wireless Power Transfer System Subject to Input Nonlinearity and Communication Delay.” IEEE Transactions on Power Electronics 38(11): 14776–14787.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Author(s)
This work is licensed under a Creative Commons Attribution 4.0 International License.
Articles published in the Journal of Advanced Multidisciplinary Studies (JAMS) are licensed under a Creative Commons Attribution 4.0 International License (CC BY 4.0), unless otherwise stated. Authors retain copyright of their work and grant JAMS the right of first publication.
