Waterfront and submarine retaining structures are normally exposed to catastrophic seepage conditions under the effect of tidal and occasionally heavy rainfall effect, resulting in a decreased passive earth thrust and thus the higher risk of instability of retaining structures. To examine the effect of seepage flow on the magnitude and distribution of passive earth thrust, this paper assumes a composite curved-planar failure surface and presents a modified method of passive earth pressure considering the seepage flow effect. The flow field and pore pressure are firstly solved by the two-dimensional (2D) Laplace equation using the Fourier series expansion. The effective reaction force acting on the composite failure surface is then obtained using a modified K & ouml;tter equation. Compared to conventional methods based on limit equilibrium, the present method facilitates a straightforward assessment of both the magnitude and distribution of passive earth thrust without the prior assumption of the application point. The outcomes highlight that the passive earth thrust decreases with the ratios of permeability coefficients. The greater effective friction angle and a smaller ratio of permeability coefficients result in the lower application point of the passive earth thrust.

The excavation of the deep foundation pit of subway station may cause excessive deformation of foundation pit and retaining structure and then pose a threat to the safety of surrounding buildings and people. Therefore, it is necessary to analyze the characteristics of ground settlement and lateral displacement of the retaining system of foundation pit caused by deep foundation pit excavation in the Guangzhou composite stratum. Based on 28 subway station projects in Guangzhou, this paper analyses the monitoring data in the process of deep foundation pit excavation and reveals the deformation characteristics of subway deep foundation pit in the Guangzhou composite stratum. The research results can provide data support for the excavation scheme design and environmental control of similar deep foundation pit projects. The results show that: (1) The final deformation of the foundation pit at Guangzhou Metro Station is predominantly within the range of 5 to -15 mm. Monitoring points with settlement values exceeding 30 mm constitute the smallest proportion, while only a limited number of measuring points exhibit surface uplift. The observed ground uplift can be attributed to two primary causes: basement heave and the infiltration of grouting slurry outside the pit. (2) The maximum ground settlement of foundation pit increases with the increase of excavation depth. When the aspect ratio of foundation pit is greater than 15, the maximum ground settlement has an obvious positive linear relationship with it. The insertion ratio of foundation pit retaining structure in the Guangzhou area is mainly concentrated in 0.30-0.59, with an average value of 0.42, and the maximum ground settlement gradually decreases with the increase of insertion ratio. (3) The maximum lateral displacement of the retaining structure of the foundation pit accounts for the largest proportion in the range of -20-25 mm. The maximum lateral displacement of the retaining structure increases with the increase of the excavation depth and the length-width ratio of the foundation pit. The maximum lateral displacement of the retaining structure decreases with the increase of the insertion ratio. There is an obvious skirting phenomenon in the granite residual soil foundation pit. Attention should be paid to and the insertion ratio should be appropriately increased in the project. (4) The maximum ground settlement caused by deep foundation pit excavation of a subway station in the Guangzhou area is 0.99-1.90 times of the lateral displacement of foundation pit retaining structure.

One of the major challenges for a numerical modeler/practitioner designing a sheet-pile retaining structure shoring waterfront regions under dynamic loading is to reasonably capture its seismic response solely on appropriate soil constitutive model calibration. This is particularly important as the numerical modelers usually do not have access to the dynamic centrifuge testing or large-scale 1G experimental facilities for the purpose of full-scale FE model validation based on the in-situ conditions. This paper emphasizes the significance of the consideration of initial static shear during the model calibration phase on the blind numerical predictions involving a sheet-pile wall retaining a submerged backfill. For this purpose, a strain space multiple mechanism model was initially calibrated against cyclic direct simple shear (CDSS) tests performed on the Ottawa F-65 sand under different initial states. The strain space multiple mechanism model was able to replicate the essential features in the cyclic soil response under the presence of initial static shear such as the non-occurrence of liquefaction and mobilization of shear strain limited to the compression side. The calibrated soil constitutive model employed within the FLIP-ROSE FE program was reasonably able to blind-predict the measured response from the centrifuge experiments. In particular, the measured seismic lateral displacements experienced by the retaining wall were excellently captured. Finally, this study signifies the role of the stress reversal and non-stress reversal cyclic loading scenarios in the elemental calibration on the extent of permanent deformation experienced by a sheet-pile wall under earthquake excitations having different characteristics. Overall, it is observed that calibration of a soil constitutive model by incorporating the essential features of the stress reversal and nonstress reversal unsymmetrical cyclic loading may result in the appropriate FE predictions for the sheet-pile wall deformation mechanism when subjected to earthquake excitations with different characteristics.

During previous medium intensity earthquakes, several cantilevered retaining structures shoring waterfront areas experienced large deformations or even collapsed. This was in contrary to the caisson type quay walls which performed better even during the very strong 1995 Kobe earthquake. Within this realm, recent studies have highlighted that saturated backfill adjacent to retaining structures may not fully liquefy and has a strong potential for shear-induced dilation mechanics owing to the presence of substantial static shear. However, despite these negative excess pore pressures and absence of a full liquefaction state in the backfill, cantilevered retaining walls may still experience large deformations. In this paper, a deformation mechanism is proposed for an embedded cantilever retaining wall supporting a submerged backfill made of Ottawa F-65 sand using a geotechnical centrifuge experiment. The dynamic response of the soil-structure system was measured, and the intra-cyclic mechanics were investigated. The entire earthquake duration of 20 s was divided into three different phases. In phase I (comprising of initial 6 s), the retaining wall experienced sliding deformations owing to its inertia with negligible soil straining at the backfill. Under the application of subsequent seismic pulses in phase II, large suction drops in excess pore pressures were observed during the translation of the retaining wall toward the backfill, which caused the negative excess pore pressure to exceed the hydrostatic value. However, the temporary release of the suction drops during the deformation of the wall toward the seaside resulted in significant softening as soil crosses the phase-transformation line. This ultimately resulted in significant plasticization of the backfill, and the wall initiated a rotation type deformation mechanism with a pivot point located near its base. Intra-cyclic observations revealed an increase in the magnitude of the phase transition in excess pore pressures, which contributed to the increased accumulated soil straining along the backfill. Owing to this, the wall experienced increasing rotations with the application of subsequent cycles in phase II (until 14 s) until the passive resistance in front of the wall was fully mobilized. However, a sudden catastrophic collapse of the wall could be avoided owing to the re-dilation mechanism, during which the soil again underwent a phase transformation and generated negative excess pore pressures on the completion of the wall translation toward the seaside. With the reduction in the amplitude of the applied cycles toward the end of shaking (phase III), the effective stress path in front of the wall moved away from the origin, and the mobilized passive stress reduced, which eventually resulted in the retaining wall achieving a stable state with no additional deformations. The proposed deformation mechanism highlights that a state of full liquefaction is not a necessary prerequisite for an embedded cantilever retaining wall to experience significant deformations, and an understanding of suction mechanics during excess pore pressure generation is critical.