| The power supply safety detection and monitoring system(6C system)plays an important role in evaluating the state of the catenary and guiding the maintenance of the catenary status to ensure the safe and stable operation of the electrified railway.With the increase of the operating mileage of electrified railways and the frequency of electric locomotives in China,the power supply operation and maintenance work is becoming increasingly heavy,so the analysis depth and efficiency of power supply detection data are put forward higher requirements.In the long-term operation of the catenary,many defects and accidents are gradually evolved from minor diseases of catenary components,and with the change of the state of the parts,the relevant catenary parameters will also change.Based on the 1C dynamic detection data,combined with the historical dynamic detection data,this paper studies the change of the equipment state by locating the location where the pull-out value parameter changes slightly.When mining catenary detection data,it is inevitable to correct the mileage deviation of detection data at different times on the same line.Considering that the contour points at the joint of the anchor section have strong characteristics,first,mathematical methods such as difference and fitting are used to describe and identify the contour points,achieving segmentation of the detection waveform,and then preliminary matching of the anchor section is carried out based on mileage information.Then,based on the layout pattern of the pull-out value data within the anchor segment,the pull-out value feature at the location of the positioning point in the anchor segment is used as the segmentation node to achieve feature principal point matching of the detection data within the anchor segment.Finally,the mileage correction of dynamic detection data between adjacent positioning points is achieved through interpolation,achieving the goal of multiple detection data alignment.On this basis,the difference between the pull-out values of the two dynamic detection waveforms is used as a reference quantity for analyzing the small changes in the pull-out values of the contact line in the anchor segment.By analyzing the change characteristics of the pull-out value difference and the change trend of the pull-out value difference at adjacent anchor points,a pull-out value change recognition algorithm based on sliding window is constructed,and the window where the pull-out value changes in the anchor segment is preliminarily located.Combined with the proximity and relative position of the difference between the pull-out value at the center anchor point in the window and the pull-out value difference of other anchor points in the window,the premises where the pull-out value changes are accurately located.The identification sample of small change in pull-out value is constructed by using dynamic detection,road bureau line parameter adjustment and daily accumulation of defect cases,and the identification algorithm of small change of pull-out value is verified.This method is used to calculate multiple high-speed railway lines,and the location of the anchor point where the identified part of the pull-out value changes is reviewed on site,the state before and after the change is compared,the reasons for the change of the pull-out value are analyzed,and the effectiveness of the identification method is confirmed. |
