Influence of dynamic soil-structure interaction on the building response to ground vibration

In the built environment, ground-borne vibration is produced by environmental sources such as road and railway traffic or construction and industrial activities. These vibrations may lead to malfunctioning of sensitive equipment, discomfort to people, and, at very high levels, structural damage. Dynamic soil-structure interaction plays a crucial role in the prediction of the response of structures to ground-borne vibration. This PhD thesis examines dynamic soil-structure interaction in a deterministic and a probabilistic setting, considering parametric uncertainty on the subsoil conditions. The emphasis is on the frequency range of interest for railway induced vibration. The availability of experimental data that can be used to calibrate and validate numerical models is essential. In this work, experimental data are used to develop and calibrate a numerical model for the prediction of railway induced vibration in an existing building. The influence of modeling simplifications and uncertain subsoil conditions is investigated. In a separate case study, the influence of heterogeneous subsoil conditions and different foundation types on the structural response is examined. The uncertainty on the subsoil conditions has a significant influence on the response of buildings to ground vibration and especially at higher frequencies. Therefore, this uncertainty should be taken into account for robust response predictions. Finite element updating is commonly used to improve the accuracy of response predictions of numerical models. However, dynamic soil-structure interaction is usually neglected in finite element updating which might lead to modeling errors adversely affecting the accuracy of the response predictions. Therefore, a methodology is developed to compute the modal characteristics of structures considering dynamic soil-structure interaction. The sensitivities of the modal characteristics are analytically derived in order to be used in gradient based optimization algorithms. The methodology is applied to calibrate the numerical model of a test structure specifically designed to study dynamic soil-structure interaction effects. As dynamic soil-structure interaction affects the modal characteristics of structures, considering its effects in vibration-based finite element updating can reduce the modeling error and, therefore, can improve the accuracy of the response predictions.status: publishe