Due to the planning of the subway route, it is difficult to avoid crossing soft soil site conditions at subway stations. The seismic response of subway station structures is closely related to the surrounding soil site. In this paper, centrifuge shaking table tests were designed and carried out for subway station structures at three typical soft soil sites (all-clay site, liquefiable interlayer site, and fully liquefiable site). The test results are as follows. The structure is most severely damaged in all-clay site, while the damage is low in liquefiable interlayer site and fully liquefiable site. For liquefiable sites, site liquefaction results in a lower soil-structure stiffness ratio. Thus liquefiable interlayer site and fully liquefiable site provide a natural seismic isolation system for structures compared to all-clay site. The limits of the inter-story drift ratio of the structure were used to evaluate the post-earthquake performance stages of the model structure in the three sites. In all-clay site, the structure is in the immediately operational stage after the loading condition of 0.1g and 0.32g, and in the reparable operational stage after the loading condition of 0.52g and 0.72g. In the liquefiable interlayer site and full liquefiable site, the underground structure is in the normal operational stage after the loading condition of 0.1g and in the immediately operational stage after the loading condition of 0.32g-0.72g.

Coral sand, as a geological material for foundation filling, is widely used for reclamation projects in coral reef areas. The coral sand is characterized by a wide grain size distribution. A series of centrifuge shaking table tests were conducted to explore the seismic response of a shallow buried underground structure in saturated coral sand and coral gravelly sand. The emphasis was placed on comparing the similarities and differences in the dynamic behavior of the underground structure at the two sites. The responses of excess pore pressure, acceleration, displacement, and dynamic soil pressure of the structure were analyzed in detail. The results indicated that the underground structure in coral sand had a significant influence on the development of excess pore pressure in the surrounding soil, but this effect was not evident in coral gravelly sand due to well-drained channels. Liquefaction was observed in the soil layer around the structure in coral sand, but it did not occur in coral gravelly sand. In coral sand, the liquefaction of the soil layer at the bottom of the structure caused a significant attenuation in the acceleration of the structure. Compared to coral gravelly sand, the acceleration response of the soil layer near the bottom of the underground structure was higher in coral sand. During the shaking, the displacement pattern of the structure in coral gravelly sand was slight subsidence-slight upliftsignificant subsidence, while it exhibited a significant uplift in coral sand. The maximum dynamic soil pressure distribution on the structural sidewalls presented a trapezoidal distribution, and the dynamic soil pressure had a strong connection with the development of excess pore pressure in the surrounding soil.

To investigate the effect of site specificity on the seismic response of underground structures, two centrifuge shaking table tests under adverse soil site conditions were carried out, considering the case of structures placed in saturated clay sites and saturated sand sites, respectively. Through the comparison of typical test results, the seismic response law of underground structures and soil sites in adverse soil sites are investigated and the damage patterns of underground structures under adverse soil sites are attempted to be revealed. The test results show that the column is the seismically weak member of the subway station structure, and its top and bottom is the earthquake damage-prone position of the member. Underground structures are more severely damaged in all- clay sites than in liquefiable sites. The liquefiable site becomes a natural vibration isolation system for underground structures. With the increase of earthquake amplitude, the liquefaction degree of the saturated sandy soil layer increases, and the increase in horizontal displacement for liquefiable sites is increasing compared to all-clay sites. Although the liquefaction of the saturated sand layer resulted in large horizontal displacements at the site, the structure did not experience large inter-story drifts compared to the all-clay site, which may have been caused by the reduction in the soil-structure stiffness ratio due to the liquefaction of the sand layer.

In conventional designs, the pile and pile cap are typically considered rigid connections. However, this type of connection experiences a concentrated force during earthquakes, leading to frequent damage at the pile heads. To mitigate pile head damage, semirigid pile-pile cap connections are proposed. Centrifuge shaking table tests were conducted to investigate the seismic response of the superstructure-pile foundation system. Two layers of Toyoura sand, including a moderately dense upper layer and a denser bottom layer, were used as the foundation soil. The superstructure was simplified as lumped masses and columns with two different heights and periods. The foundation consisted of a 3 x 3 group of piles. Rigid and semirigid pile-pile cap connections were evaluated. The experiments investigated the effect of connection type on the distribution of bending moments in the piles and analyzed the acceleration and displacement responses of the superstructure under different pile-pile cap connections. According to the results, semirigid connections reduced the peak bending moment at the pile head by 50-70 %, especially for low-rise superstructure cases. The influence depth of the connection type on the pile bending moment reaches approximately 10 times the pile diameter. For low-rise superstructure cases, semirigid connections slightly reduced the natural frequency of the superstructure, leading to a decrease in the superstructure acceleration during earthquakes with a short dominant period. The semi-rigid connections reduce the rotation of foundations but promote the translational displacement of foundations. For the mid-rise superstructure cases, semirigid connections reduce the translational displacement and increase the rotational displacement of the foundation. These experiments provide insights into the seismic performance of superstructure-pile foundation systems with different pile-pile cap connections and can serve as a reference for seismic design in similar engineering practices.

When an underground structure passes through a liquefiable soil layer, the soil liquefaction may pose a significant threat to the structure. A centrifuge shaking table test was performed to research the seismic response of underground structures in liquefiable interlayer sites, and a valid numerical model was obtained through simulation model test. Finally, the calibrated numerical model was used to perform further research on the influence of various distribution characteristics of liquefiable interlayers on the seismic reaction of underground structures. The key findings are as follows. The structure faces the most unfavorable condition once a liquefiable layer is located in the middle of the underground structure. When a liquefiable layer exists in the middle of the structure, the seismic reactions of both the underground structure and model site will increase with the rise of the thickness of the liquefiable interlayer. The inter-story drift of the structure in the non-liquefiable site is much smaller than that in the liquefiable interlayer site. The inter-story drift of the structure is not only associated with the site displacement and the soil-structure stiffness ratio but also closely associated with the slippage of the soil-structure contact interface under the condition of large deformation of the site.