A soil structure-interaction analysis of tall buildings

The earthquake records obtained from the sites with instrumented structures and free-field accelerations have provided an unprecedented opportunity to evaluate the effect of soil-structure interaction (SSI) on the seismic response of structures and to verify the developed procedures of analytical studies. This study was divided into two parts, the verification of sub-structural procedure and parametric studies. The building used in this study is a thirty-story reinforced concrete structure supported by pile foundation. It was instrumented during the Loma Prieta Earthquake, 1989. The SSI models were developed using SAP2000 and DRAIN-2D. The p-y curve procedure was utilized to define the load-deflection relationship of the pile groups.; The verification includes the dynamic response analyses of the SSI model with the free field ground motion and near field ground motion so that the effect of the kinematic interaction can also be investigated. The analytical results were compared with the recorded seismic response.; In the parametric studies, the impact of soil condition to the seismic response of the building was evaluated. The analyses were conducted on two SSI models. One of the models was developed based on the existing site condition whereas the other was modeled based on a soft soil profile. In addition, the effect of relative stiffness between the superstructure and the foundation were examined by comparing the dynamic response of a SSI model with stiffer superstructure with the SSI model based on existing building.; The results of these analyses indicated that the dynamic response of the building system highly depended on the ground motion. If the near field effect causing by the wave scattering etc. can be properly incorporated in the ground motion excitation, the sub-structural procedure can reasonably predict the dynamic response of the building. For the building having a massive foundation system, the resonance was more important than inertial interaction because the SSI modal shapes resided in the high frequency range. Furthermore, soil condition did not impact the dynamic response significantly unless the modification of the ground motion induced by change of the soil was accounted.