AI Framework Detects Hidden Power System Issues

This paper introduces a novel learning-based framework to identify and classify hidden contingencies in power systems, such as undetected protection malfunctions. This is significant because it addresses a critical vulnerability in modern power grids where standard monitoring systems may miss crucial events. The use of machine learning within a Stochastic Hybrid System (SHS) model allows for faster and more accurate detection compared to existing methods, potentially improving grid reliability and resilience.

• •Proposes a Learning-based Stochastic Hybrid System (LSHS) framework.
• •Focuses on detecting unobservable contingencies in power systems.
• •Employs machine learning for rapid and accurate contingency categorization.
• •Demonstrates improved detection speed and accuracy compared to existing approaches.