JZUS - Journal of Zhejiang University SCIENCE

Frontiers of Information Technology & Electronic Engineering

Accepted manuscript available online (unedited version)

Black-box adversarial attack on deep reinforcement learning-based PID controller for load frequency control

Author(s): Wei WANG¹, Zhenyong ZHANG^1, ², Xin WANG², Xuguo JIAO^3, ⁴
Affiliation(s): ¹State Key Laboratory of Public Big Data, Guizhou University, Guiyang 550025, China ²Key Laboratory of Computing Power Network and Information Security,Qilu University of Technology (Shandong Academy of Sciences), Jinan 250353, China ³School of Information and Control Engineering, Qingdao University of Technology, Qingdao 266061, China ⁴State Key Laboratory of Industrial Control Technology, Zhejiang University, Hangzhou 310058, China
Corresponding email(s): zhangzy@gzu.edu.cn
Key Words: Adaptive controller; Deep reinforcement learning; Load frequency control; Adversarial attacks

Share this article to： More <<< Previous Paper \|Next Paper >>>

Wei WANG¹, Zhenyong ZHANG^1,², Xin WANG², Xuguo JIAO^3,⁴. Black-box adversarial attack on deep reinforcement learning-based PID controller for load frequency control[J]. Frontiers of Information Technology & Electronic Engineering,in press.https://doi.org/10.1631/FITEE.2401021

@article{title="Black-box adversarial attack on deep reinforcement learning-based PID controller for load frequency control",
author="Wei WANG¹, Zhenyong ZHANG^1,², Xin WANG², Xuguo JIAO^3,⁴",
journal="Frontiers of Information Technology & Electronic Engineering",
year="in press",
publisher="Zhejiang University Press & Springer",
doi="https://doi.org/10.1631/FITEE.2401021"
}

%0 Journal Article
%T Black-box adversarial attack on deep reinforcement learning-based PID controller for load frequency control
%A Wei WANG¹
%A Zhenyong ZHANG^1
%A²
%A Xin WANG²
%A Xuguo JIAO^3
%A⁴
%J Frontiers of Information Technology & Electronic Engineering
%P
%@ 2095-9184
%D in press
%I Zhejiang University Press & Springer
doi="https://doi.org/10.1631/FITEE.2401021"

TY - JOUR
T1 - Black-box adversarial attack on deep reinforcement learning-based PID controller for load frequency control
A1 - Wei WANG¹
A1 - Zhenyong ZHANG^{1
A1 -}²
A1 - Xin WANG²
A1 - Xuguo JIAO^{3
A1 -}⁴
J0 - Frontiers of Information Technology & Electronic Engineering
SP -
EP -
%@ 2095-9184
Y1 - in press
PB - Zhejiang University Press & Springer
ER -
doi="https://doi.org/10.1631/FITEE.2401021"

Abstract
Chinese Summary
Academic Network
Reviewer Comment

Abstract: Load frequency control (LFC) is usually managed by traditional proportional integral derivative (PID) controllers. Recently, deep reinforcement learning (DRL)-based adaptive controllers have been widely studied for their superior performance. However, the DRL-based adaptive controller exhibits inherent vulnerability due to adversarial attacks. To develop more robust control systems, this study conducts a deep analysis of DRL-based adaptive controller vulnerability under adversarial attacks. First, an adaptive controller is developed based on the DRL algorithm. Subsequently, considering the limited capability of attackers, the DRL-based LFC is evaluated under adversarial attacks using the zeroth-order optimization (ZOO) method. Finally, we use adversarial training to enhance the robustness of DRL-based adaptive controllers. Extensive simulations are conducted to evaluate the performance of the DRL-based PID controller with and without adversarial attacks.

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference

Open peer comments: Debate/Discuss/Question/Opinion

<1>

- Go to

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference