Understanding how buried structures respond to cyclic loading is crucial for designing earthquake-resistant underground infrastructure. The complexity of the dynamic response of these underground structures is accentuated when accounting for their interaction with adjacent buildings. This study focuses on the impact of the lateral distance between a shallow rectangular tunnel and an adjacent residential structure on dynamic soil-structure interaction, employing two dynamic centrifuge tests. Particle Image Velocimetry technique was adopted to observe the developed mechanisms of liquefied soil deformation, tunnel uplift, and foundation settlement. Results show the pronounced effects on soil deformation above the tunnel crown with a reduction in the tunnel-building separation distance. The distribution of excess pore pressure along the tunnel lining will be presented. Tunnel manifested additional lateral movement and reduced uplift when the adjacent building was in close proximity. A decrease in the tunnel-building distance contributed to a reduction of building rotation, primarily due to the significant mitigation of non-uniform settlement under the footing. Overall, the soil deformation mechanism around the tunnel and the building will be shown to change depending on the separation distance between them.

It is important to determine the dynamic behaviour of underground structures under cyclic loading for the seismic underground structural design. The dynamic response of such underground structures is further complicated when considering the interaction with surface buildings. This paper presents a series of dynamic centrifuge tests and 2D numerical modelling to investigate the dynamic response of a shallow cut-and-cover rectangular tunnel and the nearby building in the liquefiable sandy ground. Dynamic soil responses such as the wave propagation and excess pore pressure are successfully captured in both centrifuge testing and numerical modelling. Tunnel uplift, building settlement and soil deformation are determined by the particle image velocimetry (PIV) technique. Results show the existence of the nearby building can cause significant effects to the tunnel lateral movement and tunnel rotations. The tunnel floatation mechanism is also discussed with a simplified vertical force equation. In addition, the presence of the buried tunnel causes non-uniform settlement distribution along the building range. The comparison of the experimental, numerical building settlement with the analytical and empirical estimations proves the limitation of these methods in considering the building interaction with other structures.