International Journal of Geomechanics

One of the most crucial tasks in the design, control, and construction of urban deep excavations is ensuring the safety of the existing underground infrastructure. Deformation and settlement created by excavation may damage the adjacent tunnels. In this study, the stability of an existing triple tunnel in relation to the construction of an adjacent deep excavation is evaluated by numerical simulation using both the discrete-element method (DEM) and the finite-element method (FEM). A deep excavation supported by the retaining wall and five levels of strutting system was created adjacent to an existing triple tunnel. The excavation’s width and depth were 30 and 16 m, respectively. In both discrete-element (DE) and finite-element (FE) simulations, the horizontal spacing of the triple tunnel wall relative to the retaining wall (SH) is varied between 3 and 35 m, while vertical spacing of the triple tunnel’s crown from the ground surface (SV) is changed from 4.8 to 32 m. The results indicated that at a certain value of SV and with increasing the SH, the horizontal displacement of the wall decreases. The variations in the triple tunnel position significantly affected the settlement pattern. In addition, the results showed that the maximum vertical displacement occurred at the middle tunnel crown, while the lowest value of the maximum vertical displacement was found at the crown of the right tunnel. At a certain value of the vertical displacement, the wall horizontal displacement is deduced by increasing in the SH value.

A numerical method is proposed to analyze the long-term structural behavior of concrete face rockfill dams (CFRDs), considering fluid–solid coupling and the spatial variability of model parameters. Initially, a method was explored for calculating the wetting deformation of rockfill within a CFRD, considering an effective stress framework and fluid–solid coupling. Subsequently, the wetting model was implemented into a FEM subroutine. Next, volumetric and axial creep curves were derived from triaxial creep tests conducted at varying porosities. Based on these test results, an enhanced creep model with higher accuracy was formulated and integrated into a finite-element calculation program. Furthermore, regression models were developed to assess the relationship between rockfill compaction quality and improved creep model parameters. The Particle Swarm Optimization (PSO)-Kriging method was employed to estimate the dam's global compaction quality. Moreover, spatial estimation of improved creep model parameters and refined assignment of FEM parameters were achieved. This refinement led to an enhanced simulation of CFRD, accounting for the wetting effect of dam materials and the spatial variability of model parameters under fluid–solid coupling. Finally, verification using a specific engineering case study demonstrated that the wetting effect of rockfill caused by impoundment and the spatial variability of creep model parameters can substantially enhance calculation accuracy. Consequently, the proposed method significantly contributes to predicting the long-term deformation of CFRD and ensuring the service safety of antiseepage concrete faces. In addition, it provides a more dependable technical method for analyzing the structural performance of CFRDs throughout their entire life cycle.

The kinematic interaction between piles under seismic loading has been extensively studied from analytical, experimental, and numerical perspectives. Of note, within numerical modeling, the majority of the existing literature relies on simplified approaches for characterizing the soil–pile interaction, which leads to the requirement for more reliable and comprehensive research. In this paper, using FLAC3D, the seismic response of the soil–pile system was investigated with a set of fully nonlinear three-dimensional (3D) numerical analyses in the time domain. This model simulated the soil strength and stiffness dependency on the stress level and soil nonlinear behavior under cyclic loading. The Mohr–Coulomb (M–C) constitutive model described the soil’s mechanical behavior, which was used with additional hysteretic damping to suit the dynamic behavior. In the framework of a parametric study, the effects of loading frequency on the response of a soil–pile system that was subjected to seismic loading were studied. The results showed that the pile response and soil characteristics, as well as the natural frequency mode of the system’s dynamic behavior, are strongly affected by the frequency of the seismic loading. Therefore, the bending moment and lateral displacement along the length of a pile increase as the loading frequency approaches the natural frequency of the system. In addition, when the loading frequency reaches a threshold value far from the fundamental frequency of the system, the effect of loading frequency on the soil–pile system response becomes negligible. In addition, the relationship between the pile diameter and maximum pile bending moment at different loading frequencies is affected by the soil properties.

Static and dynamic stability analyses are crucial components of engineering practice for ensuring the stability of phosphogypsum (PG) tailings dams. Within this context, the liquefaction analysis under seismic loads plays a pivotal role in ensuring the dynamic stability of PG tailings dams. Under seismic loads, the complicated dynamic response of the tailings dam has a significant effect on the stability of the tailings dam. In this study, considering the support of bedrock on the tailings dam, we perform a three-dimensional (3D) numerical analysis of the Zhuyuan PG tailings dam located in the city of Yichang, China, using FLAC3D. In order to consider the influence of the drainage system on the burial depth of the phreatic line, we consider a complete drainage system in the 3D model, which creates favorable drainage conditions for the tailings in its vicinity. The fully nonlinear numerical analysis method based on the Finn constitutive model and P2PSand model is employed to assess the tailings’ liquefaction potential under seismic loading. We also utilize the built-in FISH language in FLAC3D to judge, track, and label (visualize) current and past liquefaction zones in two groups, Liquefy-n and Liquefy-p, to emphasize the importance of past liquefaction zones for the dynamic stability of tailings dams. The results suggest that the P2PSand model demonstrates a stronger capability than the Finn model in capturing the development of excess pore pressure and the soil liquefaction trend over repeated cycles of loading. The liquefied PG is mainly located in the area below 5 m from the phreatic line, rather than in the area between 0 and 5 m from the phreatic line, where excess pore pressure can be quickly dispersed into the surrounding areas once it reaches the phreatic surface. Additionally, we perform a stability analysis using the strength reduction technique based on the postliquefaction residual shear strength to determine the factor of safety.