Modelling of track sound radiation

In most situations the dominant source of noise from the railway system for conventional speeds is rolling noise. This is radiated by the wheels, the rails which dominate the important mid-frequency region between 400 and 2000 Hz, and, at low frequencies, also the sleepers. The acoustic properties of the rails, sleepers and ballast are investigated in this thesis.The sound radiation of a rail in close proximity to a ground (both rigid and absorptive) is predicted by the boundary element method (BEM) in two dimensions (2D). Results are given in terms of the radiation ratio for both vertical and lateral motion of the rail, when the effects of the acoustic boundary conditions due to the sleepers and ballast are taken into account in the numerical models. Allowance is made for the effect of wave propagation along the rail by applying a correction in the 2D modelling. The numerical predictions of the sound radiation from a rail are verified by comparison with experimental results obtained using a 1:5 scale rail model in different configurations.The sound radiation from the sleepers has been calculated using a three-dimensional boundary element model including the effect of both reflective and partially absorptive ground. When the sleeper flexibility and support stiffness are taken into account, it is found that the radiation ratio of the sleeper can be approximated by that of a rigid half-sleeper. When multiple sleepers are excited through the rail, their sound radiation is increased as they form a composite source. This effect has been calculated for cases where the sleeper is embedded in a rigid or partially absorptive ground. It is shown that it is sufficient to consider only three sleepers in determining their radiation ratio when installed in track. At low frequencies the vibration of the track is localised to the three sleepers nearest the excitation point whereas at higher frequencies the distance between the sleepers is large enough compared with the acoustic wavelength for them to be treated independently. Measurements on a 1:5 scale model railway track are used to verify the numerical predictions with good agreement being found for all configurations.Not only can ballast absorb noise to some extent due to the gaps between the ballast particles but it can also vibrate and reradiate noise during a train pass-by. Experiments have been performed using 1:5 scale ballast to investigate this. The basic properties of the ballast, as a porous material, are measured initially, in particular the flow resistivity and porosity. The modelling of the ballast absorption is then implemented based on the corresponding measured parameters and this is compared with the measured absorption. The effects of ballast absorption on the rail and sleeper radiation are also measured, and are compared with the numerical predictions. Moreover, the vibration of the ballast is obtained experimentally. The influence of the ballast vibration on the sleeper radiation is then estimated and shown to increase the noise radiation below 200 Hz.Finally, the sound radiation from the whole track is predicted, and compared with the corresponding measured results on the 1:5 scale model. The sound radiation models for the rail and the sleeper, are used with the TWINS software to give revised predictions of the track sound radiation. Prediction differences are shown between the original TWINS and the updated models