Shaking table test and numerical research on seismic response of a CFST arch bridge for different soil grounds

Abstract

Severe damages or failures of arch bridges on different soil grounds have been reported during previous strong earthquakes. The influence of different soil grounds on the seismic response of arch bridges is unclear. In this study, three large-scale shaking table tests were conducted to investigate the seismic response of a model bridge abutment of a concrete-filled steel tubular (CFST) arch bridge, which was set inside a model soil box at a reduced scale of 1/90. Grounds with three types of soil were simulated: silty clay, clay-gravel layered, and bare gravel soil. The accelerations and displacements of the model soils and the sliding displacement and settlement of the model bridge abutment were evaluated through the tests. The test results showed that the soil types significantly affected the seismic response of the soil and bridge abutment system, and the spectral characteristics of input waves significantly affected the acceleration amplitude ratio (AAR) of the bridge abutment; however, the changing trend of the AAR was usually related to the corresponding soil types. The soil grounds exhibited a remarkable filtering effect along the soil heights under seismic excitations, and the filtrations of the frequency components of the input waves for different soil grounds shows significant difference. Furthermore, the dynamic shear stress–strain curves showed that the silty clay soil remained in the nonlinear behavior development stage under strong seismic excitations, while the bare gravel soil remained in the linear behavior development stage. The energy dissipation capacity of the bare gravel soil ground could not be utilized. The test results in this study can be used for seismic risk assessment of different soil grounds, to guide the optimal seismic design of arch bridges for different soil grounds.