Distributed Anomaly Detection in Modern Power Systems: A Penalty-based Mitigation Approach

Abstract

The evolving landscape of electric power networks, influenced by the integration of distributed energy resources require the development of novel power system monitoring and control architectures. This paper develops algorithm to monitor and detect anomalies of different parts of a power system that cannot be measured directly, by applying neighboring measurements and a dynamic probing technique in a distributed fashion. Additionally, the proposed method accurately assesses the severity of the anomaly. A decision-making algorithm is introduced to effectively penalize anomalous agents, ensuring vigilant oversight of the entire power system's functioning. Simulation results show the efficacy of algorithms in distributed anomaly detection and mitigation.

Tasks

Reproductions