Satellite and UAV-Assisted Transmission Line Fault Prediction Based on GNN

With the rapid expansion of the power industry, ensuring the secure and stable operation of transmission lines has become critical to the reliable functioning of the entire power system. Traditional inspection approaches, primarily relying on manual patrols, periodic maintenance, and localized sensor monitoring, face significant challenges in addressing wide-area faults, sudden equipment failures, environmental variations, and the high randomness introduced by emerging loads such as new energy vehicle charging. To address these limitations, this paper proposes a fault prediction framework for transmission lines based on Graph Neural Network (GNN), leveraging the coordinated sensing capabilities of low earth orbit (LEO) satellites and unmanned aerial vehicle (UAV). By integrating satellite remote sensing and UAV inspection data, a multimodal representation of transmission line structures is constructed. The GNN model then fuses spatially and temporally heterogeneous information to capture the spatiotemporal evolution patterns of faults. Experimental results demonstrate that the proposed method significantly improves fault prediction accuracy and localization performance while reducing false alarm rates, offering both theoretical insights and practical value for condition monitoring and proactive maintenance in smart grids.