Method of Prediction of Track Technical Condition Based on the Use of Irregularity Size Distribution Function

摘要

The article describes the properties and application of the original multipoint method for predicting the values of track geometric parameters and their changes over time. The method uses a database containing the results of previous measurements of track geometry as input data. To calculate predicted values of track geometric parameters, we applied autoregressive model to the input data by minimizing (least squares method) the forward and backward errors. Prediction results are presented in the form of irregularity size distribution function (ISDF). ISDF shows the cumulative length of each-size track irregularity within a track segment. The method is invariant to the length of the time interval between past measurements. The outcome of the analysis is the set of predicted values of the ISDF at the end of the prediction interval. Use of ISDF allows to predict track condition in future periods with high accuracy. For surface and alignment defects, for example, the average prediction error for defect sizes s＞ |3| mm does not exceed 0.66 m for a 100 m track segment.