Unmanned aerial vehicles in microgrid defense

Abstract

In the evolving landscape of microgrid cybersecurity, this paper introduces a groundbreaking approach using unmanned aerial vehicles (UAVs) integrated with a deep neural network‐robust optimization (DNN‐RO) framework to defend against sophisticated false data injection attacks (FDIAs). Our research pioneers a dynamic UAV‐based defense model, meticulously engineered and simulated across a 50 square kilometer virtual microgrid. The UAVs leverage cutting‐edge path loss models and strategic energy management to optimize their deployment and operational efficiency. Our extensive simulation trials reveal compelling outcomes: a reduction in detection latency by over 50%, classification accuracy improved to 94.7%, and a streamlined response time that robustly counters cyber threats. Furthermore, the operational deployment of UAVs achieves significant cost reductions, showcasing not only the model's enhanced security capabilities but also its economic benefits. This study not only marks a significant advance in microgrid protection strategies but also sets a new standard for integrating UAV technology in critical infrastructure defense, offering scalable and economically feasible solutions for future‐proofing energy systems against cyber threats.