Abstract

Railway transport is widely promoted as the most environmentally friendly and safe means of transport in terms of energy consumption, CO2 and exhaust atmospheric emission, and reported accident levels. A major environmental challenge to increased rail transport is the generation of low frequency vibration (1-80 Hz) and re-radiated noise (16-250 Hz) in buildings. Vibration and noise can cause annoyance to residents and disrupt sensitive equipment, ultimately threatening main European rail corridors and hindering development of new railways. Legal limits and best-practice guidelines exist to minimize vibration disturbance, but the complexity of the urban environment means that accurately predicting vibration levels remains an extremely difficult task, especially for new-built situations. During the past decades, considerable progress was made regarding the development of semi-analytical and numerical models for the prediction of railway induced vibration in the built environment. These include, for example, 2.5D and periodic coupled finite element – boundary element (FE-BE) or FE-PML models of the track and soil, as well as 3D FE-BE models for buildings accounting for dynamic soil-structure interaction (SSI). These models are used to design vibration mitigation measures and to perform environmental impact studies. Accurate numerical prediction of railway induced vibration in buildings remains very challenging, however, given the complexity of the coupled dynamic SSI problems involved, the wide frequency range of interest, the large number of determining parameters and their associated level of uncertainty. We will illustrate these challenges by comparison of model predictions with experimental results obtained during a large scale in situ measurement campaign in a three-storey reinforced concrete building located in close proximity of a ballast track on embankment that is operated by freight and passenger trains. Industry experiences a strong need to conduct extended parametric studies (e.g. track type, soil characteristics, building typology) and to optimize vibration mitigation measures (track, transmission, building) in a robust way. This is illustrated by means of two examples: vibration mitigation of an urban tramway on soft soils by means of a (floating) slab track supported by piled foundations, and the use of seismic metamaterials as a vibration mitigation measure on the transmission path. These applications all require detailed vibration prediction models that are fast to run. We therefore also investigate the potential of model order reduction techniques to tackle the “curse of dimensionality” and considerably speed-up state-of-the-art prediction models.