Pantograph Spark Fault Detection using YOLO

Pantograph-catenary is now the dominant form of current collection for modern electric trains because they can be used higher voltages. Faults in pantograph-catenary systems threaten operation and safety railway transportation. They need to continuously monitored controlled maintain safe transport. Pantograph may damaged as a result extreme weather conditions which affect its normal operation, leading failure pantograph overhead contact line systems. Poor between causes thermal erosion wire. When pantographs are exposed air, could deteriorate due electrochemical reaction with environment since made metals. Movement catenary lines high crosswinds has been found cause wire trapped pantograph. There serious issue regarding quality images generated by video monitoring system on high-speed often shows inconsistencies faults. The application traditional image processing deep learning techniques have unable meet requirements spark detection. In this paper, algorithm proposed detect sparks Specifically, YOLOv3 model counter problem that algorithms to. results very large sample data show efficiency real-time performance method, meets detection railway. Keywords : High-speed pantograph; Spark detection; Deep learning; YOLOv3; DOI: 10.7176/ISDE/12-3-02 Publication date: September 30 th 2021