Research On Prediction Model Of Corrective Maintenance Needs Related To Railway Track Geometry

With the continuous development of railway construction in the direction of high speed and heavy load,research on the prediction model of corrective maintenance needs and the prediction of the spatiotemporal position of corrective maintenance related to railway track geometry plays an important guiding role in reducing the times and cost of corrective maintenance related to track geometry,and for ensuring the normal operation of the railway line.Based on the analysis of the research status of corrective maintenance related to railway track geometry at home and abroad,with the management requirements of corrective maintenance related to railway track geometry,the prediction model of corrective maintenance needs related to railway track geometry was established in this paper.The model mainly includes:(1)Combining with the management standard of corrective maintenance related to railway track geometry,the probability model of corrective maintenance related to track geometry was established to statistically analyze the potential law of corrective maintenance related to track geometry,and quantify the probability of the corrective maintenance.(2)Using the grey system theory and method and the railway infrastructure gridding management theory and method,the grey interval prediction model and the grey compositional data model were established separately for the track quality index(TQI)and the track geometry irregularity index structure of the 200 m track segment.Combining with the results of the track quality index prediction model and the track geometry irregularity index structure prediction model,the track geometry condition of each track segment was precisely predicted.(3)Combining with the probability model of corrective maintenance related to track geometry and the track geometry condition prediction model of each segment,the corrective maintenance needs related to track geometry in the time period between the next two adjacent inspection times were quantified and predicted,and the spatiotemporal positions of the corrective maintenance were predicted.Finally,a total of 48 track inspection car historical data from September 2014 to December 2018,maintenance operation data from 2014 to 2018 and the basic account data of the Shenmu-Shuozhou railway was collected,by participating in the Shenshuo Railway Branch Company project "Shenshuo Railway Branch Company Operation Integrated Business Management System" undertaken by Beijing Jiao tong University.And the above data was used to verify the effectiveness of the prediction model of corrective maintenance needs related to railway track geometry established in this paper.By comparing the predicted spatiotemporal position of corrective maintenance related to track geometry based on the prediction model with the actual corrective maintenance of the Shenmu-Shuozhou railway,the results show that the prediction model of corrective maintenance needs related to railway track geometry established in this paper is effective.