International Journal for Numerical and Analytical Methods in Geomechanics最新文献

Accurate calibration of discrete element simulation parameters (characteristic parameters) for unstructured pavements is a prerequisite for model application, particularly in analyses such as tire-pavement contact mechanics. Focusing on soil pavements, experiments such as the soil angle of repose were conducted, using the angle of repose as the response value. A fusion screening method for significant characteristic parameters of unstructured pavements was proposed using Plackett–Burman experiments, the Spearman correlation coefficient, and the gradient boosting decision tree algorithm (GBDT). The steepest ascent experiment determined the optimal range of the significant characteristic parameters, and a polynomial kernel-based support vector regression (poly-SVR) model was introduced to obtain their optimal values. Based on this, a discrete element model for the unstructured pavement was established, and its validity was demonstrated through compaction tests. Results showed an average relative error of 0.55% between simulated and measured soil repose angles, demonstrating the feasibility of the significant characteristic parameter fusion screening method and the poly-SVR model for identifying optimal parameter values. Additionally, the mean relative error between the measured and simulated soil compaction values was determined to be 7.40%, confirming the validity of the established pavement model. The presented calibration method for discrete element simulation parameters of unstructured pavements can serve as a theoretical reference for related research.

This paper investigates the stress–strain relationships and volumetric behaviors of expanded polystyrene (EPS) under compression and monotonic/cyclic shearing, considering the effects of vertical stress (σ_v) and EPS density (ρ_EPS). The experimental results demonstrate that under monotonic shearing, EPS exhibits strain hardening, and the resulting shear-induced axial strain (ε_a) significantly exceeds the creep-induced ε_a under sustained σ_v. The apparent shear strength (τ_f) is governed by equivalent cohesion (c), which increases linearly with ρ_EPS, while the equivalent friction angle (φ) remains insensitive to ρ_EPS. Under cyclic shearing, symmetric hysteresis loops form, with secant shear stiffness (G) increasing with ρ_EPS and σ_v but decreasing with shear strain amplitude (γ_a), the equivalent damping ratio (D) shows opposite trends. Axial strain accumulates nonlinearly during cycling, stabilizing after about four cycles under fixed γ_a but increasing continuously under multistage loading. An extended hypoplastic model is proposed, incorporating two key advancements: the characterization of shear-induced volumetric contraction and a shear strain-dependent equivalent cohesion to describe damage accumulation. The model accurately simulates the stress–strain relationships and axial strain evolution of EPS with different densities and stress levels under both monotonic and cyclic shearing, providing a robust tool for analyzing EPS in geotechnical applications.

As the utilization of shallow and medium-depth underground spaces approaches saturation, the future development of urban underground space will increasingly shift toward deeper strata. The geological composition of these underground spaces primarily consists of soil layers. The ground pressure acting on underground structures is closely related to ground deformation. However, traditional loosening earth pressure theory fails to account for the effect of ground deformation on ground pressure. This study presents an elastic–plastic solution for the deformation pressure of deep-buried circular tunnels under non-uniform stress fields. To derive the elastic–plastic solution, a new method is proposed for constructing the stress function. This method enables the determination of the elastic stress field of a tunnel in an infinite domain under gravity. The obtained elastic–plastic solution effectively captures the evolution of ground pressure during tunnel deformation. The validity of the elastic–plastic solution is verified through finite element analyses and model tests. In addition, the proposed method is compared with the loosening earth pressure theory. The results indicate that as the C/D ratio (C and D denote the tunnel burial depth and diameter) increases, the ground pressure calculated by the loosening earth pressure theory decreases, whereas the pressure obtained from the proposed method increases. Under finite deformation conditions, the loosening earth pressure theory tends to underestimate the tunnel load.

Clay liners are widely used in waste containment systems to prevent groundwater contamination. This study presents transient analytical solutions for solute transport coupled with consolidation in a semi-infinite domain. No closed-form solution exists for this configuration despite its common use in theoretical analyses. The solution is derived using the Laplace transform and validated against existing models and experimental data. It provides explicit expressions for solute concentration and pore water pressure, capturing the combined effects of deformation, seepage, and diffusion. Based on this framework, new formulas for clay liner thickness are proposed for two breakthrough criteria (I and II). These formulas enable mechanical load to be incorporated into the design. Results show that the coupled effects of solute transport and consolidation significantly accelerate solute migration, as consolidation-induced pore water flow enhances advective transport. Compared with finite boundary conditions, a semi-infinite boundary slows solute migration. The required liner thickness increases with mechanical loading, with Criterion I exhibiting a substantially higher growth than Criterion II, although the growth rate decreases with increasing load for both criteria. Under a mechanical loading of 400 kPa, the thickness predicted by Criterion I is 25.7% greater than that predicted by Criterion II. In the landfill case study, the proposed methodology confirms the adequacy of the existing liner and demonstrates potential cost savings of 27.0% and 49.5% under Criteria I and II, respectively. This work provides an efficient tool for evaluating solute transport coupled with consolidation and offers practical guidance for clay liner design optimization in engineering applications.

Existing suffusion models inadequately characterize soil grading parameters, failing to quantify particle changes and comprehensively map soil property evolution during erosional processes. This paper presents a suffusion model for soil–rock mixtures, enabling quantitative assessment of mass changes across different particle groups. The model integrates four key computational modules: the seepage control equation, critical conditions for soil particle movement, particle grading equations, and porosity relationships. The feasibility of the proposed model for predicting mass changes in particle groups during the suffusion of soil–rock mixtures was subsequently validated through two seepage tests. Finally, a comprehensive stability analysis of the soil–rock mixture slope was conducted using the refined gradation-based suffusion model. This analysis evaluated volume water content, pore water pressure, particle gradation, and factor of safety. The validations between experimental results and the analysis method confirm the validity and robustness of the formulated suffusion model. This study realizes the quantitative analysis of each particle group during the process of suffusion of soil–rock mixture, and further predicts the stability of soil–rock slope during the suffusion process.

Expanded polystyrene (EPS) geofoam is increasingly used in retaining structures as a compressible inclusion to reduce lateral earth pressures. However, the complex interactions among soil, EPS, and wall components—together with the limitations of analytical and numerical approaches—pose significant challenges in accurately predicting lateral pressures on rigid retaining walls with geofoam inclusions. This study introduces an intelligent hybrid Sparrow Search Algorithm–Machine Learning (SSA–ML) framework to overcome these deficiencies through autonomous hyperparameter optimization, interpretability enhancement, and uncertainty quantification. Six ML models—Generalized Neural Regression (GNRR), Support Vector Machine (SVM), Extreme Learning Machine (ELM), Decision Tree (DT), Random Forest (RF), and Multilayer Perceptron (MLP)—were optimized using SSA based on a comprehensive database compiled from experimental and numerical studies. Model performance was systematically assessed through K-fold cross-validation and comparative analysis. The kernel density estimation (KDE) method was applied to generate probabilistic interval predictions, thereby quantifying the inherent uncertainty in the modeling process. Furthermore, the SHapley Additive exPlanations (SHAP) approach was employed to interpret the influence and directionality of key parameters on predicted lateral pressures. Results demonstrate that the proposed SSA–MLP model achieves superior predictive accuracy and robustness compared to other benchmark models, while maintaining clear physical interpretability. This hybrid, interpretable, and uncertainty-aware framework provides a reliable data-driven tool for analyzing lateral earth pressures on soil–EPS–wall systems and offers new insights for the design and optimization of compressible inclusion retaining walls.

The bedding planes in layered rock masses are critical structures that control strength and stability. This study systematically investigates the strength and deformation characteristics of layered rock under different confined pressures using the FDEM method. Based on the global embedded cohesive element in Abaqus, the differentiated characterization of bedding planes and rock matrix was achieved, and numerical specimens of layered rock samples were constructed. To comprehensively reveal the mesoscopic failure mechanisms of layered rock under different confined pressures, the precise discrimination of crack dynamic thresholds based on material properties was achieved through Python. Simultaneously, the damage distribution of layered rock is obtained, which shows the spatial distribution of shear and tension damage intuitively. The results show that the compressive strength exhibits U-shaped with the bedding inclination. The compressive strength, elastic modulus, and peak strain demonstrate an approximately linear relationship with the increase in confined pressure. The new discriminant threshold obtained by the study can effectively distinguish the damage modes, and under different confined pressures, the proportion of crack types and the proportion of cracks at the matrix and bedding plane are significantly dependent on the bedding inclination. Meanwhile, the failure mechanism of the samples can be effectively characterized by the spatial distribution characteristics of shear and tension damage. Finally, the applicability of the proposed method is verified by comparing the results with published literature.

Finite–discrete element method (FDEM) has become a widely recognised numerical method for simulating the fracturing behaviour of materials under various loading conditions. However, the substantial computational cost of three-dimensional (3D) FDEM has led to a marked imbalance: extensive research exists on two-dimensional (2D) FDEM, while studies on 3D FDEM remain limited. This study investigates the dynamic fracture behaviours of rocks under multiaxial static and dynamic coupled loads, utilising a self-developed 3D FDEM parallelised based on a general-purpose graphics processing unit (GPGPU). The 3D FDEM incorporates both intrinsic and extrinsic cohesive zone models (ICZM and ECZM) as well as various contact interaction algorithms to facilitate robust simulation of a full-scale triaxial Hopkinson pressure bar (Tri-HB) testing system. Dynamic uniaxial and biaxial compression (UC and BC) tests within the Tri-HB framework are modelled with different combinations of cohesive zone models and contact algorithms, and the results are compared against each other and against laboratory experiments reported in the literature. The simulations demonstrate that the 3D FDEM with all these models and algorithms can reasonably capture the stress wave propagations in both metal bars and rocks as well as the primary dynamic fracturing behaviours of rocks under the coupled static and dynamic loads. However, computing accuracy and efficiency vary across model combinations. Overall, the 3D FDEM with the ECZM achieves the highest accuracy and efficiency.