Explore the vast range of titles available.

The axial force in assembly steel struts with servo systems is a critical indicator of stability in foundation pit support systems. Due to its high sensitivity to temperature variations and direct influence on the lateral deformation of the foundation pit enclosure structure, accurate prediction is essential for safety monitoring and early warning. This study proposes a novel method for predicting the axial force in assembly steel struts with servo systems based on a spatiotemporal adaptive network. The method begins by feeding historical axial force data from multiple steel struts into an LSTM network to extract temporal sequence features. A self-attention mechanism is then employed to capture the global dependencies within the axial force data, enhancing the feature representation. Concurrently, a convolutional neural network (CNN) is utilized to extract local spatial features. Additionally, excavation depth and excavated soil stratification data are processed through convolutional operations to derive stratification-related features. Subsequently, the temporal and spatial features of axial force are fused with stratification-related features derived from excavation data and further refined through a CNN, enabling more accurate predictions. Validation using deep foundation pit data from a metro station in Zhejiang Province demonstrated the method’s reliability and improved performance across multiple metrics compared to the existing approaches.

Under the increasingly frequent extreme rainfall events, deep excavations are subjected to complex coupled effects of excavation-induced disturbance and rainfall infiltration, making it essential to clarify the deformation mechanisms of surrounding soils and retaining structures. In this study, a typical metro foundation pit in Zhengzhou was selected as the research object. By integrating field monitoring data with ABAQUS finite-element simulations, the deformation response and mechanical evolution of the pile–soil system under the combined influence of staged excavation and extreme rainfall infiltration were systematically investigated. The results indicate that rainfall infiltration leads to significant pore-water pressure accumulation, while the downward migration of the wetting front continuously reduces the effective stress and shear strength of the mid-to-deep soils. This process weakens the passive resistance at the pile toe and induces the downward movement of the primary deformation zone. The bending moment distribution of the retaining piles evolves from a “coexisting positive–negative moment” pattern to a “positive-moment dominant” mode, and the horizontal displacement at the pile toe changes from negative to positive, revealing a coupled mechanism involving deep-soil softening and arching loss. A mechanical chain of rainfall infiltration, pore-pressure evolution, stress redistribution, and pile–soil deformation is established to explain how extreme rainfall amplifies excavation-induced effects through dual pathways of mid-to-deep soil softening and accelerated seepage. Based on these findings, a prevention framework of source reduction, structural enhancement, and process control is proposed to improve the safety and resilience of deep excavations under extreme climatic conditions.

When a foundation pit is hydraulically connected with surroundings, the dewatering inside the excavation would both induce water-level decline and enclosure wall deflection, which together cause ground settlement outside the excavation. However, the current studies have not fully revealed the settlement behaviour under the combined actions of the above two factors; meanwhile, the individual effect of the two factors on the ground settlement is still indistinguishable. In this study, in situ pumping test and numerical simulations were both conducted to ascertain the above issues. Specifically, a fluid–solid coupling numerical model was developed to simulate a practical foundation pit dewatering test; measured ground settlement and groundwater drawdown were adopted to validate the numerical model; then, a series of numerical simulations were performed to revel the characteristics of ground settlement due to the combined actions of groundwater drawdown and wall movement; on this basis, the individual impacts of the two factors on the ground settlement were separated. Results show that the settlement ratio caused only by enclosure wall movement (ηb) varies in the range of 2.5–43%, depending on the pumping location, pumping time and hydraulic connectivity in strata; ηb is overall greater during the pumping of phreatic aquifer compared to the pumping of confined aquifer, while with the pumping time elapsed, ηb would both decrease apparently regardless of the pumping location.

The hardening soil model with small strain stiffness is a valuable tool for predicting the deformation of support structures during the excavation phase of construction projects. The stiffness parameters of the model, which are dependent on technically complex and costly tests or estimated through specific proportionate, may exhibit some discrepancy between the analyzed results and the project monitoring data in certain aspects. In light of the findings from the conducted analyses and studies, the new concepts of reference in-situ overburden pressure and reference in-situ void ratio are proposed with the objective of enabling the determination of the essential parameters required for HSS model through the utilization of the results of the current geotechnical tests, which are of high popularity and economy. Additionally, the article offers recommendations for determining other parameters, providing a summary of a systematic approach to determining the necessary parameters for the hardening soil model with small strain. Ultimately, a comparison and verification of the finite element method results of a deep excavation project with the monitoring data demonstrated that the research outcomes exhibited sufficient numerical analysis accuracy and practical applicability.

The deformation of existing piles adjacent to deep excavations was governed by strongly coupled pile-soil interaction. To overcome the limitations of traditional approaches, this study developed a unified analytical model based on the total potential energy principle and the variational method. By formulating the strain energy of the pile and soil together with the external work of vertical load and excavation-induced lateral earth pressure, the coupled governing equation for pile displacement was rigorously derived and solved analytically. A laboratory-scale excavation test with four staged conditions was conducted to validate the proposed model. The predicted lateral displacement profiles exhibited strong agreement with the measured data. At key depths, the absolute prediction error generally remained within 0.02–0.03 mm, and the relative error was mostly below 10%, with local maxima up to about 20% near the pile toe. These larger deviations near the pile toe were attributed to increased boundary stiffness and soil shear effects. Overall, the results demonstrated that the proposed analytical framework accurately captured the depth-dependent deformation characteristics of piles adjacent to excavations and provided a reliable theoretical basis for deformation prediction and safety evaluation in deep excavation engineering.

Most rockbursts are associated with jointed rock masses as the energy stored inside the rock is prone to abruptly release at such discontinuities. Moreover, remote blasting at working faces can trigger the brittle failure of jointed rocks at opening boundaries. In this work, a testing system combining true-triaxial compression with dynamic disturbance is employed to investigate the failure modes of cuboid jointed granite specimens. A three-dimensional tunneling-induced stress path is created in cuboid jointed granite specimens by first applying static loads to simulate the stress state encountered around an open boundary. Then, a dynamic disturbance (load) is continuously applied to the granite specimen to simulate the effect of the vibrations experienced due to remote blasting. A dynamic disturbance in the σ2 direction is found to reduce the peak strength of the specimen (σp) by approximately 5% (compared to static stress-induced failure) and promote shearing failure at the fracture surface. The major fracture surfaces form a ‘V’-shape, along with several nearby minor cracks. However, a dynamic disturbance in the σ3 direction can reduce the peak strength by 10% and accelerate the evolution of tensile failure resulting in subvertical tensile fracture. Due to the natural stiffness of the joint, the Poisson’s ratio (υ3) in the ɛ3 direction is larger than that in ɛ2 direction, highlighting the anisotropy of the deformation. The proportion of the total stored energy that is released in the jointed rock is greater than that released in intact hard rock. More strain energy is dissipated at the joint fracture. The joint provides a location at which energy can be released, which considerably increases the severity of potential rockbursts. The energy-storage limit of granite with different dip angles is also explored based on complete stress–strain curves. Post-peak granite specimens with joints dipping at 60° are found to be the most susceptible to external dynamic disturbance but release less kinetic energy. Specimens with joints that dip at 80° incorporate the maximum amount of stored energy and are not prone to rockburst when subjected to dynamic disturbance. Once a rockburst does occur in 80°-dipping jointed granite, a strong tremor will appear, however.HighlightsDisturbance in σ2 direction decreases strength by 5% and expedites shearing failure.Disturbance in σ3 direction reduces strength by 10% and accelerates tensile failure.Poisson’s ratio (ν3) in ε3-direction presents more than that (ν2) in ε2-direction.The joint dipping of granite significantly affects energy components of rockbursts.

In response to increasingly stringent safety and construction environmental impact requirements for urban underground engineering, this study investigates the performance of a steel strut servo system braced deep excavation adjacent to existing buildings, using a subway station project as a case study. A three-dimensional finite element model is established to analyze the behavior of the steel strut servo system during deep excavation, focusing on the deformation characteristics of soil and structural members, as well as the factors influencing the system’s performance. The findings indicate a strong correlation between the deformation of soil and structural members and the excavation depth, with greater deformation observed at deeper depths. When excavation is completed, the maximum and minimum vertical displacement of soil mass are 24.3 and − 5.8 mm, respectively. The maximum total displacement of buildings A and B is 3.86 and 3.82 mm, respectively. The servo system can inhibit the displacement of diaphragm wall to some extent. The maximum values of the servo area and other areas are 25.21 and 40.4 mm, respectively. The axial force of the strut is mainly pressure, with a maximum value of − 2,883.4 kN. The horizontal displacement of diaphragm wall is sensitive to the change of servo system position and servo axial force value. The deformation control effect is best when all steel struts are controlled by servo system. When the servo system is set with 2 struts, the maximum value decreases as the position of the servo system moves down. In addition, the maximum displacement decreases with the increase of the servo axis force value. This research provides valuable insights for the design optimization and construction control of similar projects.

In soft clay, deep excavations adjacent to tunnels cause complex soil–structure interactions. We conducted centrifuge tests with Particle Image Velocimetry (PIV) to simulate a staged deep-pit excavation near a model tunnel. A scaled retaining wall and tunnel lining were instrumented in a strongbox; the soil was consolidated and excavated in four stages under 60 g. PIV tracked soil and structure displacements while pore-pressure sensors recorded stresses. Tunnel position (beside vs. below the pit) and lining stiffness were varied to isolate their effects. The results reveal a shielding effect: the tunnel acts as a rigid strut that redistributes stresses and mitigates excavation-induced settlement. Surface settlement and retaining-wall deflection were lower than in a no-tunnel case. This shielding depends on tunnel stiffness and proximity: a stiffer tunnel provides greater soil restraint, whereas a flexible lining allows more movement. A tunnel close to the excavation (within roughly one to two pit depths) bears higher internal load but yields the largest reduction of far-field displacement. PIV shows soil arching: settlement above the tunnel is reduced, while heave develops at the pit base. Three characteristic uplift patterns emerge: a symmetric “hill”, a central “groove”, and an asymmetric “triangle” toward the tunnel. These patterns reflect how soil arching is altered by the adjacent tunnel and wall. We define a critical interaction depth where the tunnel’s role shifts from passive inclusion to an active structural element. When the tunnel lies in this vertical zone near the pit bottom, it markedly alters stress paths and uplift geometry. By highlighting the tunnel’s dual role—reducing wall deformation while sustaining higher internal stress—and by categorizing uplift shapes and the depth threshold of interaction, this study advances understanding of tunnel–excavation interaction. These contributions (quantified shielding metrics, uplift-pattern classification, and the critical-depth concept) provide a basis for design and deformation prediction in deep excavations near tunnels.

In urban concentrated area, the disturbance caused by construction affects significantly the sustainability of adjacent existing structures. It is essential to capture the mechanical response of existing structures to adjacent deep excavation. The objective of this paper is to investigate the displacement and internal force behavior of elevated bridge piles (BP) subject to influence of deep excavation. A three-dimensional finite element model is established by taking the project of a deep excavation near elevated bridge as an example. The numerically calculated results agree well with the measured data, which verifies the established numerical model. On the basis of this model, the influence of deep excavation on the mechanical characteristics of adjacent piles is captured. The results show that the displacement, bending moment, and shear force of piles are sensitive to the excavation depth. Their magnitudes increase with the increase of excavation depth. When the excavation is completed, the maximum displacements of piles in horizontal direction and vertical direction are 2.3 mm and 10.05 mm, respectively. The maximum bending moment is 1,140.8 kN·m. The maximum and minimum shear forces are 1,206 kN and -2,282 kN, respectively. The piles are mainly under pressure. The maximum pressure is -13,116 kN. The axial force is not sensitive to the depth of excavation. The deformation and internal force of piles exhibit obvious spatial distribution characteristics, and the closer the distance to the middle of the long side of the deep excavation, the greater the value. The research results have a positive effect on the optimization of related engineering structures and the promotion of sustainable development in urban concentrated area.

This paper proposes a convolution neural network (CNN) based prediction method for concrete diaphragm wall (CDW) deflections. CNN algorithm is modified for processing the CDW deformation data collected from in-situ measurement in both time and space dimensions, and capable of making dynamic prediction based on the extracted spatiotemporal features of wall deflection. The proposed method is validated through investigating a project of deep excavation in Suzhou, China. The predicted results show excellent agreement with field measurement and yield mean absolute errors of 0.86 mm and 1.55 mm for nowcasting and forecasting tasks, respectively. Three prevailing algorithms in time series prediction, namely, back propagation neural network, long short-term memory and autoregressive integrated moving average, are conducted for comparison. The results illustrate that the CNN outperforms the other algorithms in terms of accuracy and execution time. Therefore, the proposed CNN model is the most suitable for CDW deflection prediction, and can provide reasonable references for construction safety management on site.