Adversarial Attacks on Deep Learning-Based False Data Injection Detection in Differential Relays

2025-06-24

AI Summary

This work exposes severe adversarial vulnerability in deep learning–based false data injection attack (FDIA) detection models deployed in line current differential relays (LCDRs) of smart grids. To exploit this weakness, we propose the first adversarial attack framework tailored to differential protection’s remote current measurements, which injects imperceptible perturbations into sampled current waveforms to achieve stealthy evasion and induce spurious tripping. We conduct the first systematic evaluation of adversarial robustness across MLP, CNN, LSTM, and ResNet architectures on FDIA detection, revealing attack success rates exceeding 99.7%. Furthermore, we design an active defense mechanism integrating adversarial training, achieving substantial robustness improvement without compromising fault detection accuracy (>99.5%). Our findings provide both theoretical insights and practical solutions for securing deep learning–enabled protection systems in smart grids.

Technology Category

• Machine Learning: Adversarial Learning & Robustness
• Computer Vision: Adversarial Attacks & Robustness
• Multiagent Systems: Adversarial Agents

Application Category

• User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systems
• Security and Privacy: Large-scale security measurements
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

The application of Deep Learning-based Schemes (DLSs) for detecting False Data Injection Attacks (FDIAs) in smart grids has attracted significant attention. This paper demonstrates that adversarial attacks, carefully crafted FDIAs, can evade existing DLSs used for FDIA detection in Line Current Differential Relays (LCDRs). We propose a novel adversarial attack framework, utilizing the Fast Gradient Sign Method, which exploits DLS vulnerabilities by introducing small perturbations to LCDR remote measurements, leading to misclassification of the FDIA as a legitimate fault while also triggering the LCDR to trip. We evaluate the robustness of multiple deep learning models, including multi-layer perceptrons, convolutional neural networks, long short-term memory networks, and residual networks, under adversarial conditions. Our experimental results demonstrate that while these models perform well, they exhibit high degrees of vulnerability to adversarial attacks. For some models, the adversarial attack success rate exceeds 99.7%. To address this threat, we introduce adversarial training as a proactive defense mechanism, significantly enhancing the models' ability to withstand adversarial FDIAs without compromising fault detection accuracy. Our results highlight the significant threat posed by adversarial attacks to DLS-based FDIA detection, underscore the necessity for robust cybersecurity measures in smart grids, and demonstrate the effectiveness of adversarial training in enhancing model robustness against adversarial FDIAs.

Problem

• Adversarial attacks evade deep learning-based FDIA detection in smart grids
• Existing deep learning models are highly vulnerable to adversarial FDIAs
• Adversarial training enhances model robustness against FDIA attacks

Innovation

• Fast Gradient Sign Method for adversarial attacks
• Adversarial training enhances model robustness
• Evaluates multiple deep learning models' vulnerabilities