Abstract Summary
Ground motion induced by human activities related to construction has been a classical topic of great interest. Damage to nearby structures due to settlement, disturbance of occupants, jeopardy of serviceability and structural integrity are common effects of excessive vibrations. Pile driving is one of the main activities that involves the preceding risks and a number of studies has been conducted with a view to understand and minimize those risks. Currently, the environmental impact of vibrations induced by anthropogenic activities is drawing further attention. Infrastructure projects in urban areas have been traditionally the focus of such studies. However, at present the disturbance induced by dynamic sources in other habitats gains significant interest, with a view to endangerment of various species. A major activity in the field of offshore wind, that pertains to this case, is the installation of offshore foundations, i.e. monopiles. To estimate the level of environmental vibrations induced by such activities and assess the necessity of mitigation measures, improvement of the available predictive tools is necessary. In this paper, a numerical study is performed to investigate the effect of a non-linear pile-soil interface on the resulting driving-induced soil motion. In particular, a non-linear 3-D axisymmetric pile-soil interaction model that has been developed for vibratory installation analysis is employed, that is capable of describing the pile penetration process and has been benchmarked against field data. The model is comprised by a thin cylindrical shell (pile), a linear elastic layered half-space (soil) and a history-dependent frictional interface. The wave motion in the soil is captured accurately, by modelling the layered soil half-space via the Thin-Layer Method (TLM) coupled with Perfectly Matched Layers (PMLs). The numerical solution of the coupled problem is sought via a hybrid frequency-time scheme based on sequential application of the Harmonic Balance Method (HBM). The numerical results are compared with the respective ones from a linear model that neglects the non-linear pile-soil contact. Comparisons are drawn between the two approaches in terms of ground motion magnitude and frequency content, showcasing the effect of pile penetration inclusion on the resulting wavefield in the soil medium.
