Influence of the vehicle model on the prediction of the maximum bending response of simply-supported bridges under high-speed railway traffic

This investigation intends to contribute to the problem of train-bridge interaction in high-speed railway lines. For design purposes, most often a numerical model based on constant moving loads is used to reproduce the traffic action exerted on the structure. Since no vehicle–bridge interaction is considered, this simple model may overestimate the bridge transverse displacements and accelerations when resonance occurs, particularly in simply-supported spans. According to Eurocode 1 the reduction of the bridge response due to this phenomenon may be included either by means of a coupled vehicle–bridge analysis, for which several vehicle models can be employed, or by considering an additional amount of damping which depends on the bridge span. In this investigation the two approaches are addressed and the bridge response at resonance provided by both techniques is thoroughly compared; the comparison is carried out through various examples where the interaction phenomenon may have practical importance in the verification of the serviceability limit state of acceleration for ballasted tracks. Using a nondimensional formulation, the key parameters that determine the reduction of the bridge peak response in terms of accelerations and displacements due to train-bridge interaction are brought to light. The results of a subsequent extensive sensitivity analysis are presented and discussed. The main purpose is to decide whether, taking into account all potential train-bridge configurations, a conservative approach in the prediction of the response reduction caused by interaction can still be adopted.