Gas Leakage Detection Method Based on YOLOv8 and Optical Flow

Ultra-high voltage gas-insulated switchgear (GIS) widely uses SF<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</inf> as insulation, yet gas leakage poses safety risks and environmental hazards. This study proposes an SF<inf xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">6</inf> leakage detection method integrating YOLOv8-based object detection with optical flow analysis. Experimental results demonstrate that the hybrid approach significantly enhances detection accuracy and efficiency for GIS equipment, reduces manual inspection workload and time costs, and minimizes environmental risks. The proposed method advances intelligent and digitalized maintenance practices in power systems by enabling rapid, automated leakage monitoring.