Charging Ahead: A Hierarchical Adversarial Framework for Counteracting Advanced Cyber Threats in EV Charging Stations

TL;DR

This work tackles the security challenge of advanced cyber threats in EV charging stations where malicious EVs may falsify State of Charge (SoC) to gain charging priority. It introduces a hierarchical adversarial DRL framework (HADRL) comprising an adversarial DRL agent that generates sophisticated stealthy attacks and a DRL-based intrusion detection system (IDS) trained on these attacks using LSTM/Transformer architectures and PPO-based training. The key contributions are: (i) a method to synthesize realistic, hard-to-detect attack data via DRL, (ii) a robust DRL-IDS capable of detecting unseen attacks with high accuracy, and (iii) empirical validation on a real-world-like EV taxi dataset showing near-perfect detection metrics and resilience to varying training conditions. The findings imply that combining adversarial data generation with DRL-based detection substantially improves EVCS resilience to evolving cyber threats, supporting safer deployment of smart charging infrastructure.

Abstract

The increasing popularity of electric vehicles (EVs) necessitates robust defenses against sophisticated cyber threats. A significant challenge arises when EVs intentionally provide false information to gain higher charging priority, potentially causing grid instability. While various approaches have been proposed in existing literature to address this issue, they often overlook the possibility of attackers using advanced techniques like deep reinforcement learning (DRL) or other complex deep learning methods to achieve such attacks. In response to this, this paper introduces a hierarchical adversarial framework using DRL (HADRL), which effectively detects stealthy cyberattacks on EV charging stations, especially those leading to denial of charging. Our approach includes a dual approach, where the first scheme leverages DRL to develop advanced and stealthy attack methods that can bypass basic intrusion detection systems (IDS). Second, we implement a DRL-based scheme within the IDS at EV charging stations, aiming to detect and counter these sophisticated attacks. This scheme is trained with datasets created from the first scheme, resulting in a robust and efficient IDS. We evaluated the effectiveness of our framework against the recent literature approaches, and the results show that our IDS can accurately detect deceptive EVs with a low false alarm rate, even when confronted with attacks not represented in the training dataset.

Figures (3)

• Figure 1: The system model with ($K$) EVs, charging station, and charging controller.
• Figure 2: Convergence of the stealthy DRL agent with various random seeds.(a) LSTM-based DRL Model (b) Transformers-based DRL Model.
• Figure 3: DRL-IDS convergence ((a) LSTM-Based model, (b) Transformers-based model) for different Learning rate values.