Korrugazioa iragartzeko eredu eraginkor baten garapena eta haren aplikazioa metro-linea erreal batean

Abstract

This paper develops a computationally efficient time/space domain model that combines vertical and lateral dynamics of the track and the wheelset to predict rail corrugation as a function of track and vehicle parameters. The model is obtained by transforming a frequency domain model to the time domain using the Rational Fraction Polynomials method. With this method, the fit of a transfer function associated with a receptance is obtained. Using multi-objective genetic algorithms, the error in the fitting of the receptances describing the vertical and lateral dynamics of both the track and the mounted axle is minimised, in addition to ensuring the stability of the system. In this model, the tangential stresses are obtained using the FASTSIM algorithm. The model is validated by means of experimental measurements of receptaces and corrugation carried out on a metro line. The control point selected for this validation is one of those that has experienced the greatest corrugation growth over the 5 years of the study. The wear results obtained using the model show a great similarity with the experimental results.