A multi-module robust method for transient stability assessment against false label injection cyberattacks

The success of deep learning in transient stability assessment (TSA) heavily relies on high-quality training data. However, the label information in TSA datasets is vulnerable to contamination through false label injection (FLI) cyberattacks, resulting in degraded performance of deep TSA models. To address this challenge, a Multi-Module Robust (MMR) TSA method is proposed to rectify the supervised training process misguided by FLI attacks in an unsupervised manner. In MMR, a supervised classification module and an unsupervised clustering module are alternately trained to improve the clustering friendliness of representation leaning, achieving accurate clustering assignments. By leveraging the clustering assignments, we construct a training label corrector to rectify the injected false labels and correct the misguided supervised classification, thereby improving the performance of deep TSA models. However, there is still a gap on accuracy and convergence speed between MMR and FLI-free deep TSA models. To narrow this gap, we further propose a human-in-the-loop training strategy, named MMR-HIL. In MMR-HIL, potential false samples can be detected by modeling the training loss with a Gaussian distribution. From these samples, the most likely false samples and most ambiguous samples are selected and re-labeled by a TSA expert-guided annotator and then subjected to penalized optimization, aimed at improving accuracy and convergence speed. Extensive experiments indicate that MMR and MMR-HIL both exhibit powerful robustness against FLI attacks. Moreover, the contaminated labels can be effectively corrected, demonstrating superior resilience of the proposed methods.