The noise radiating from a steel bridge can be much greater than the pass-by noise levels of trains at-grade. This is attributed to a special phenomenon known as the steel bridge-borne noise, which occurs when shocks and vibrations due to moving train loads are transferred onto the bridge. To investigate the effectiveness of elastomer mats for reducing vibration and noise in rail transit steel bridges, a frequency domain prediction model is established and calibrated through a field experiment. In this model, the train and track interactions were investigated analytically, and the receptance technique was introduced to determine the force transmitted from a multi-layer track structure to the bridge deck. The local vibrations of the bridge components were solved numerically using a hybrid shell/beam finite element model based on a harmonic response analysis. The bridge vibrations were regarded as the source of noise radiation by considering each bridge component as a rectangular flat plate noise source. There was good agreement between the calculated and measured results in terms of both the magnitude and frequency dependence. The mechanisms of vibration transmission and noise radiation were investigated using numerical simulations. In addition, the vibration and noise reduction effects are discussed by numerically comparing a track system with and without an elastomer mat.

Integrated building–bridge structures are increasingly common in high-speed railway stations, where the elevated track floor is directly subjected to train–track dynamic forces that result in excessive vibrations. The common methods of predicting the train-induced vibrations in large-scale IBBS are implemented in the time domain and can have prohibitively long computation times. This paper presents a frequency domain model for the vibration analysis of large-scale IBBS, with the aim of reducing the computation time while retaining sufficient accuracy. The model consists of three coupled subsystems: train, track, and IBBS. The train and track subsystems are investigated analytically, and the IBBS subsystem is solved numerically using a finite element method. The receptance technique is introduced to obtain the wheel/rail force. The force transmitted to the floor slab is treated as the vibration source of the IBBS subsystem. The simulated vibration levels in the IBBS subsystem are compared with those obtained from in situ measurements, and a good agreement is observed in terms of both magnitude and frequency dependence. The vibration responses of the IBBS subsystem at different locations of the track floor and the waiting floor are compared, and the influence of the track position is investigated. Finally, a parametric analysis is conducted with the aim of formulating anti-vibration measures, in which the carbody acceleration, rail displacement, and ballast acceleration are considered as key indicators. The force transmission, vibration transmission and IBBS vibrations are also investigated. The results indicate that using a ballast mat and enlarging the column cross-section are the two most promising measures for reducing the vibration levels.