An efficient approach for prediction of subway train-induced ground vibrations considering random track unevenness

In this paper, an efficient and robust approach is presented for the prediction of subway train-induced ground vibrations considering random track unevenness. The proposed approach is developed by a two-step approach. Firstly, the power spectral density functions of the wheel–rail interaction forces are obtained through the combination of the vehicle–track–tunnel–soil theoretical model and the pseudo-excitation method (PEM). Secondly, by assuming the geometry of the track, tunnel, and soil to be homogeneous along the track, the nonstationary random ground vibrations are solved by combining the PEM and the 2.5D finite element–perfectly matched layers method. To improve computational efficiency, an efficient wavenumber sampling scheme is proposed. In the numerical examples, the proposed approach is validated by comparing the results obtained with the published results. The influence of the train speed on the ground vibrations are also investigated, in which two types of track structure, namely the direct fixation track and floating slab track, are considered.