![]() |
ADAPTIVE MULTI-AGENT REINFORCEMENT LEARNING FOR ENHANCING SECURITY AND PRIVACY IN EV FAST-CHARGING NETWORKS FOR SUSTAINABILITY |
---|---|
รหัสดีโอไอ | |
Creator | Pannee SUANPANG |
Title | ADAPTIVE MULTI-AGENT REINFORCEMENT LEARNING FOR ENHANCING SECURITY AND PRIVACY IN EV FAST-CHARGING NETWORKS FOR SUSTAINABILITY |
Contributor | Pitchaya JAMJUNTR, Chanchai TECHAWATCHARAPAIKUL, Chutiwan BOONARCHATONG, Wattanapon CHUMPHET, Nawanun SRISUKSAI |
Publisher | Asian Interdisciplinary and Sustainability Review |
Publication Year | 2568 |
Journal Title | Asian Interdisciplinary and Sustainability Review |
Journal Vol. | 14 |
Journal No. | 2 |
Page no. | Article 6 |
Keyword | Adaptive Multi-Agent Reinforcement Learning, Electric Vehicle, Fast Charging Network, Security and Privacy, Sustainability |
URL Website | https://so05.tci-thaijo.org/index.php/PSAKUIJIR |
Website title | https://so05.tci-thaijo.org/index.php/PSAKUIJIR/article/view/279881 |
ISSN | 3027-6535 |
Abstract | Electric Vehicle (EV) adoption is rapidly increasing, necessitating robust and secure fast-charging networks. However, existing infrastructures face significant security and privacy challenges. This paper proposes an innovative approach using Adaptive Multi-Agent Reinforcement Learning (MARL) to address these issues. Our methodology involves formulating the problem within a MARL framework, designing adaptive agents that optimize security protocols while preserving user privacy. We conducted experiments in a simulated EV charging environment, demonstrating that our approach enhances security measures such as intrusion detection and privacy-preserving data handling. Key findings indicate significant improvements in network resilience and user privacy, validated through comprehensive metrics and visualization. This research contributes to advancing the understanding and application of MARL in critical infrastructure security and suggests future directions for integrating adaptive intelligence into EV charging networks for sustainability. |