The influence of source–receiver interaction on the numerical prediction of railway induced vibrations

Abstract

Abstract The numerical prediction of vibrations in buildings due to railway traffic is a complicated problem where wave propagation in the soil couples the source (railway tunnel or track) and the receiver (building). This through–soil coupling is often neglected in state-of-the-art numerical models in order to reduce the computational cost. In this paper, the effect of this simplifying assumption on the accuracy of numerical predictions is investigated. A coupled finite element–boundary element methodology is employed to analyze the interaction between a building and a railway tunnel at depth or a ballasted track at the surface of a homogeneous halfspace, respectively. Three different soil types are considered. It is demonstrated that the dynamic axle loads can be calculated with reasonable accuracy using an uncoupled strategy in which through–soil coupling is disregarded. If the transfer functions from source to receiver are considered, however, large local variations in terms of vibration insertion gain are induced by source–receiver interaction, reaching up to 10 dB and higher, although the overall wave field is only moderately affected. A global quantification of the significance of through–soil coupling is made, based on the mean vibrational energy entering a building. This approach allows assessing the common assumption in seismic engineering that source–receiver interaction can be neglected if the distance between source and receiver is sufficiently large compared to the wavelength of waves in the soil. It is observed that the interaction between a source at depth and a receiver mainly affects the power flow distribution if the distance between source and receiver is smaller than the dilatational wavelength in the soil. Interaction effects for a railway track at grade are observed if the source–receiver distance is smaller than six Rayleigh wavelengths. A similar trend is revealed if the passage of a freight train is considered. The overall influence of dynamic through–soil coupling in terms of power flow remains limited to 2 dB, but the insertion gain at particular locations can easily reach 10 dB. This is of the same order of magnitude as other sources of uncertainty in the numerical prediction of railway induced vibrations; this should hence be accounted for when performing vibration predictions.