Research On The Online Pantograph Fault Detection System Based On Image Processing

With the rapid development of urban rail transit,the requirement for the operation safety of urban rail trains is getting higher and higher.It is of great significance in ensuring pantograph and catenary safety to detect pantograph faults on-line and avoid operation accidents of urban rail train caused by pantograph faults.Aiming at three kinds of common pantograph faults,such as slide wear,center line offset and sheep horn missing,this paper studies and designs an online pantograph fault detection system based on image processing on the basis of summarizing the research results at home and abroad.The main contents of this paper are as follows:(1)The overall design of pantograph fault online detection system is carried out.On the basis of demand analysis,the overall structure of the system is designed,the working flow of the system is determined,and the components of the system are analyzed.(2)The algorithm of pantograph slide wear detection is studied.Aiming at the complex image acquisition in the scene,the skateboard is located based on the principle of line projection transformation;on the basis of traditional edge detection algorithm,the adaptive Canny edge detection algorithm is used to detect the pixel-level edge of the skateboard;by analyzing the characteristics of the skateboard edge,the sub-pixel edge detection algorithm based on cubic polynomial fitting is used to detect the sub-pixel edge of the skateboard.(3)The detection algorithms of pantograph center line offset and sheep horn missing are studied.Aiming at the center line offset fault,the image features are analyzed.The skateboard feature points and corners feature points are detected based on Hough transform and template matching,and the center line offset values are calculated according to the pantograph model on site.For the corners missing fault,considering that it belongs to the second classification problem,the corners missing detection algorithm based on HOG and SVM is used to detect whether the corners are missing or not.Judge.(4)According to the system design,the system is installed and debugged in Guanhu Station of Guangzhou Line 13,and the pantograph slide wear algorithm,center line offset and corners missing detection algorithm are tested by field data.The experimental results show that the performance of the algorithm meets the system requirements.