CRUPL: A Semi-Supervised Cyber Attack Detection with Consistency Regularization and Uncertainty-aware Pseudo-Labeling in Smart Grid

TL;DR

The paper tackles cyber-attack detection in smart grids under limited labeled data by proposing CRUPL, a semi-supervised approach that fuses consistency regularization with uncertainty-aware pseudo-labeling and curriculum learning. It employs a Temporal CNN (TempCNN) to process time-series measurements and iteratively labels unlabeled data using soft labels and adaptive, per-class thresholds, reducing confirmation bias. Empirical results on two public datasets show high detection accuracy (≥98%) and low false positive rates, with robust performance on unknown attack patterns. The method offers scalable, multi-class anomaly understanding beyond binary detection, potentially improving response strategies in cyber-physical grid systems.

Abstract

The modern power grids are integrated with digital technologies and automation systems. The inclusion of digital technologies has made the smart grids vulnerable to cyber-attacks. Cyberattacks on smart grids can compromise data integrity and jeopardize the reliability of the power supply. Traditional intrusion detection systems often need help to effectively detect novel and sophisticated attacks due to their reliance on labeled training data, which may only encompass part of the spectrum of potential threats. This work proposes a semi-supervised method for cyber-attack detection in smart grids by leveraging the labeled and unlabeled measurement data. We implement consistency regularization and pseudo-labeling to identify deviations from expected behavior and predict the attack classes. We use a curriculum learning approach to improve pseudo-labeling performance, capturing the model uncertainty. We demonstrate the efficiency of the proposed method in detecting different types of cyberattacks, minimizing the false positives by implementing them on publicly available datasets. The method proposes a promising solution by improving the detection accuracy to 99% in the presence of unknown samples and significantly reducing false positives.

Figures (5)

• Figure 1: Pseudo-labeling
• Figure 2: The Temporal CNN model used for Pseudo-labeling
• Figure 3: Pseudo-Labeling with Consistency Regularization and Curriculum Learning
• Figure 4: Power System Framework hink2014machine
• Figure 5: Performance of Proposed Method on (a) Dataset 1 and (b) Dataset 2