Information Fusion and Image Processing Based Arc Detection and Localization in Pantograph-Catenary Systems

Rail transit system represented by high-speed railway supplies power to locomotive electric equipment by connecting pantograph and catenary. The off-line phenomenon often occurs during the operation of pantograph-catenary system and generates off-line arc. Off-line arc spark causes drastic change of locomotive current and aggravates wear of pantograph slide plate and catenary, which seriously endangers the safe operation of rail transit. The automatic analysis and synthesis of dynamic image information, speed information acquired in time sequence and static track information in map database by computer simulation technology to complete the detection and geographic localization of pantograph-catenary arc defects are researched in this paper. Useful information needed is obtained by fusion of sensor information and human-derived prior information: according to the fusion of prior knowledge and image processing, a pantograph location method based on connected region and arc recognition method based on regional color feature are proposed. The degree of arc defect and catenary wear is obtained by fusion of image detection results and speed information, the severity of defect is determined quickly, and the state of arc is monitored in real time. Through the fusion analysis of arc monitoring results and highspeed rail speed and track information, the localization of arc defects in pantograph-catenary is determined, and the region of serious wear of catenary is obtained, which provides a reference for overhaul and maintenance of catenary. This method runs fast and has strong real-time performance, which meets the requirements of real-time detection and localization of pantograph-catenary arc.