Journal article · Ahead of Print article
Vulnerability Identification and Remediation of FDI Attacks in Islanded DC Microgrids Using Multi-agent Reinforcement Learning
University of Louisiana at Lafayette1
Electronics, Department of Electrical Engineering, Technical University of Denmark2
Department of Electrical Engineering, Technical University of Denmark3
Smart Electric Components, Center for Electric Power and Energy, Centers, Technical University of Denmark4
Center for Electric Power and Energy, Centers, Technical University of Denmark5
This paper proposes a novel approach to uncover deficiencies of the existing cyber-attack detection schemes and thereby to serve as a foundation for establishing more reliable cybersecure solutions, with particular application in DC microgrids. For this purpose, a multi-agent deep Reinforcement Learning (RL) based algorithm is proposed to automatically discover the vulnerable spots on the conventional index-based cyberattack detection schemes, and automatically generate coordinated stealthy destabilizing False Data Injection (FDI) attacks on cyberprotected islanded DC microgrids.
To enable a continuous action space for the trained RL agents and enhance the algorithm’s precision and convergence rate, Deep Deterministic Policy Gradient DDPG) is incorporated. Using this approach, susceptibility of a state-of-the-art detection scheme to several different coordinated FDI attacks on the distributed communication links is identified.
The proposed algorithm is also enhanced with a sniffing feature to enable maintaining the stealthy attacks even under the sudden disconnection of any of the compromised links. To address the discovered deficiencies within the index-based detection scheme, a complementary multi-agent RL detection algorithm using Deep Q-Network (DQN) is integrated, which provides a more reliable overall identification performance.
Taking into account the communication delays and load changes, the effectiveness of the proposed algorithm is verified by the experimental tests.
Language: | English |
---|---|
Publisher: | IEEE |
Year: | 2022 |
Pages: | 6359-6370 |
ISSN: | 19410107 and 08858993 |
Types: | Journal article and Ahead of Print article |
DOI: | 10.1109/TPEL.2021.3132028 |
ORCIDs: | Wan, Yihao , Mijatovic, Nenad , Dragicevic, Tomislav , 0000-0003-2895-5805 and 0000-0002-0214-6519 |