COMPUTATIONALLY EFFICIENT MODEL FOR CALCULATING THE 2.5D GREEN'S FUNCTIONS OF A TUNNEL BURIED IN A LAYERED HALF-SPACE SYSTEM

摘要

Noise and vibration pollution caused by underground railway traffic in urban areas become a major issue of concern due to their effects on the quality of life and comfort of the inhabitants. An accurate and computationally efficient model is required to predict the ground-borne vibration and to evaluate performance of countermeasures before their implementation. In this paper, a computationally efficient model to calculate vibration levels induced by underground railways buried in a layered half-space is presented. It is assumed that near-field dynamic behavior of the tunnel-soil system is only influenced by the dynamics of the tunnel itself and the layer that contains it. Thus, it is assumed the other layers and the free surface do not affect the dynamic behavior of the tunnel-soil system close to the tunnel. The two-and-a-half-dimensional (2.5D) tunnel-soil interface vibration displacements are calculated by using the Pipe-in-Pipe (PiP) model in which tunnel wall and surrounding soil are modeled using thin shell theory and elastic continuum theory, respectively. Then, the 2.5D Green's functions for a homogeneous full-space are employed to find a set of equivalent forces that can reproduce the tunnel-soil interface displacements. Finally, the far-field vibration displacements of the layered half-space are calculated by employing computationally efficient method which calculates the 2.5D Green's functions for a layered half-space based on the stiffness matrix method in Cartesian coordinates. The results are compared with the ones calculated by using a well-established model proposed by Hussein et al. (JSV, Vol. 333 (25), pp. 6996-7018, 2014) in which the stiffness matrix method in cylindrical coordinates has been used to calculate the 2.5D Green's function for a layered half-space. It is shown that not only the results of two methods are in a good agreement but also the new method is computationally more efficient in comparison with the existing one.