Monitoring, Inspection and Early Warning System in Electrical Distribution Networks Using Satellite Images

Timely identification of problems in electrical distribution networks is crucial to preventing major failures, reducing costs, and ensuring a reliable power supply. This paper presents a monitoring, inspection, and early warning system designed specifically for electrical networks, utilizing satellite imagery to complement traditional inspections. The system uses spectral indices derived from satellite images to monitor environmental factors such as humidity, vegetation, snow cover, and burned areas, offering a comprehensive view of the grid’s surroundings. Collected daily, this information detects changes that may pose risks to power lines and infrastructure. The system also allows users to include custom indices, ensuring flexibility in various environmental and network contexts. An integrated AI model estimates vegetation height from Sentinel-2 images, identifying potential risk areas where vegetation could threaten power lines. One key advantage of the system is the reduced reliance on costly, frequent manual inspections, lowering operational expenses compared to other methods like aerial photography or LiDAR scanners. Additionally, it provides early alerts to grid operators when potential issues are detected, enabling timely intervention and proactive maintenance. This improves network efficiency and reliability by enhancing the response to critical situations and facilitating preventive risk management.