Towards Robust Stability Prediction in Smart Grids: GAN-based Approach under Data Constraints and Adversarial Challenges
This addresses the challenge of ensuring availability and security in smart grids for energy management, though it is incremental as it builds on existing GAN methods for anomaly detection.
The paper tackles the problem of predicting grid stability in smart grids under data constraints and adversarial threats by introducing a GAN-based framework that generates out-of-distribution samples from stable data to detect instability without requiring unstable data, achieving up to 98.1% accuracy in stability prediction and 98.9% in adversarial attack detection with an average response time under 7ms.
Smart grids are crucial for meeting rising energy demands driven by global population growth and urbanization. By integrating renewable energy sources, they enhance efficiency, reliability, and sustainability. However, ensuring their availability and security requires advanced operational control and safety measures. Although artificial intelligence and machine learning can help assess grid stability, challenges such as data scarcity and cybersecurity threats, particularly adversarial attacks, remain. Data scarcity is a major issue, as obtaining real-world instances of grid instability requires significant expertise, resources, and time. Yet, these instances are critical for testing new research advancements and security mitigations. This paper introduces a novel framework for detecting instability in smart grids using only stable data. It employs a Generative Adversarial Network (GAN) where the generator is designed not to produce near-realistic data but instead to generate Out-Of-Distribution (OOD) samples with respect to the stable class. These OOD samples represent unstable behavior, anomalies, or disturbances that deviate from the stable data distribution. By training exclusively on stable data and exposing the discriminator to OOD samples, our framework learns a robust decision boundary to distinguish stable conditions from any unstable behavior, without requiring unstable data during training. Furthermore, we incorporate an adversarial training layer to enhance resilience against attacks. Evaluated on a real-world dataset, our solution achieves up to 98.1\% accuracy in predicting grid stability and 98.9\% in detecting adversarial attacks. Implemented on a single-board computer, it enables real-time decision-making with an average response time of under 7ms.