Validation of a vertical train–track–bridge dynamic interaction model based on limited experimental data

Abstract

This article focuses on the calibration and validation of a numerical model of a steel-concrete composite railway bridge considering the vehicle–track–structure interaction. The calibration of the numerical model of the bridge-track system involved the use of a genetic algorithm, and was based on modal parameters obtained from an ambient vibration test, which was performed with significant limitations regarding the sensors placement. The model updating results demonstrated a considerable degradation of the ballast in the interfaces between the decks, particularly in the longitudinal joint between half-decks, essentially due to the movements induced by rail traffic. The validation of the dynamic behaviour of the train-bridge system was performed based on the comparison between experimental responses obtained from a dynamic test under railway traffic and numerical responses, based on train–bridge interaction analysis. A good agreement was achieved between experimental and numerical records regarding the displacements at the supports and deck, and the accelerations at the deck. In case of the deck accelerations, the damping coefficient values adopted in the dynamic analysis, estimated from the logarithmic decrement method, were decisive for reaching the best adjustment between experimental and numerical records.