Research On Detection Algorithm Of Contact Wire Wear And Catenary Dynamic Geometric Parameters Of Rail Transit

The PCS(pantograph-catenary system)is an important component of the rail transit power supply system.The reliability of the PCS directly affects the performance and operational safety of trains.To grasp the operating state of the PCS,high-precision real-time detection of the contact wire wear and the catenary dynamic geometric parameters is required.The detection algorithm is the core of the pantograph-catenary detection system,which directly affects the accuracy and robustness of the detection results.Commonly used detection algorithms have the problem of low accuracy and poor robustness.Research and improvement of the detection algorithm can improve the accuracy of contact wire wear detection and catenary dynamic geometric parameters detection,and help improve the competitiveness of the detection products.First,the contact wire wear detection algorithm is studied.For low accuracy and poor robustness of the current contact wire wear detection system,structured light is used to detect wear data,and a contact wire wear detection algorithm based on Hough Transform is designed.The algorithm preprocesses the profile data collected by the structured light device with mean shift algorithm and cubic spline interpolation,determines the center of the contact wire according to Hough Transform,and calculates the wear based on the geometric characteristics of the contact wire.Through the detection of a 400 m line,the false detection rate of the algorithm using the Hough Transform is 15.6% lower than that of the Least Squares.In addition,the variance detected by the algorithm in this thesis is 0.0075,which is lower than the variance detected by the Least Squares of 0.0937.Experimental results show that the algorithm in this thesis is more accurate and robust.Secondly,the detection algorithm of catenary dynamic geometrical parameters is studied.To solve the problems of long calculation time and identification errors of traditional detection algorithms,a real-time detection algorithm of the conductor height and the stagger value based on image processing is designed.To avoid meaningless calculation and improve the speed of calculation,the region of interest extraction algorithm is used.To avoid too many detection lines affecting the detection results and improve the calculation speed,the LSD(Line Segment Detector)algorithm,short straight-line judgment and merging algorithm are used,and the matching conditions of contact line features are analyzed.Compared with the Hough Transform detection algorithm,the experimental results show that the standard deviation of the calculated results of the algorithm in this thesis is lower,which indicates that the algorithm of this thesis is faster and more accurate.Finally,the accuracy,robustness,and calculation speed are verified by experimental data,all of which meet the application requirements.Through the line test,the detection algorithm is applied to urban rail transit and other similar detection scenarios,including the detection scenarios of subway third rail positioning and rail contour matching.The experimental results show that the method proposed in this thesis is effective,has better robustness than traditional detection methods,and can be applied to similar detection scenarios.The contact wire wear detection algorithm and catenary dynamic geometric parameters detection algorithm proposed in this thesis can solve the problems of insufficient accuracy and robustness of commonly used detection algorithms,and realize real-time detection of contact wire wear and catenary dynamic geometric parameters.It can provide ideas for laser detection and feature extraction scenarios,and provide a reference for improving other structured light detection systems.The detection results can provide references for the intelligent maintenance of PCS.