基于多次波形匹配的高速铁路动检数据里程误差评估与修正

获取具有准确里程信息的动检车检测数据,是实现高速铁路线路的高效养护维修与分析其状态演变规律的基本前提。针对当前处理动检数据里程误差的不足,如区段内数据波形重复性差或依据单次检测数据处理误差等会造成错误修正,通过引入约束条件、动态尺度系数以识别、处理特殊区段并综合考虑多次检测数据,提出一种更可靠的里程误差评估模型,采用拉格朗日乘子法求解该模型并基于线性变换与插值方法修正里程误差,最后应用该方法编制了动检数据分析软件。结合某高速铁路动检数据研究发现:不合理的模型尺度参数会降低修正精度,建议取40~120m;在99.7%置信度下,任意两次动检数据间里程误差可控制在0.54m内;本文方法能有效处理实际工程中动检数据的里程误差问题,结合数据点标准差方法可实现快速定位线路几何状态波动明显的位置并准确评估线路养护维修作业效果。

Mileage Error Estimation and Correction for High-speed Railway Track Inspection Data Based on Multiple Data Waveform

Accurate position for track inspection data is the basic precondition to guarantee the efficient maintenance and accurate prediction for the status of high-speed railway.Limits of current research in reducing milepost errors are that errors could be wrongly corrected if section has poor repeatability in data waveform or single track inspection data is used to be the calibration reference.A more reliable milepost error estimation model was established based on the constraints introduced and dynamic scale coefficient to identify and process special sections and to comprehensively consider multiple track inspection data.The Lagrangian multiplier method was applied to solve the model.Based on the solution,milepost errors were corrected through linear transformation and interpolation method.Finally,the method was used to develop the inspection data analysis software.Based on the high-speed railway data from track inspection car,the results show that unreasonable model scale parameters can lower the correction precision and the parameters are recommended to be 40-120 m.Under the confidence of 99.7%,the differences of mileposts between any two inspection data fall in the range of 0.54m.The presented model is effective to solve the milepost error problem in practical engineering.It can rapidly identify the larger fluctuation of track geometry state and evaluate the effect of track maintenance operation.Accordingly,it is significant to guide maintenance and management of railway and ensure the safe operation.