Effect of Excess Pore Pressure on Earthquake‐Induced Displacement of Partially Saturated Sandy Soil Slopes: Flexible Sliding Block AnalysisZhang, Tong; Ji, Jian; Du, Shigui; Song, Jian; Huang, Wengui
2024 International Journal For Numerical and Analytical Methods in Geomechanics
doi: 10.1002/nag.3855
The permanent displacement of earth slopes during earthquake shaking is a key indicator for landslide hazard assessment. Previous studies mostly attempt to evaluate the earthquake‐induced displacement of dry or saturated soil slopes, while it is less common to deal with partially saturated soils. In the present study, a simplified procedure is proposed to account for the seismic‐induced excess pore pressure in slopes with partially saturated sandy soils. The effect of matric suction, suction stress, and excess pore pressure on the yield acceleration of partially saturated sandy slopes is investigated, and the coupled Newmark sliding block method, known as the flexible soil columns with dynamic shear modulus and damping ratio, is modified to estimate the seismic slope displacement. Detailed discussions are made about the effect of different degrees of saturation on the excess pore pressure ratio, yield acceleration, and slope displacement. The numerical results show that the excess pore pressure ratio tends to exponentially increase with saturation, and the change of yield acceleration and displacement with saturation can be divided into suction stress dominant and excess pore water pressure dominant stages.
Analytical Solutions for Consolidation of Soft Ground With Impervious Columns Considering Non‐Darcian FlowLi, Kuo; Lu, Mengmeng; Sun, Jinxin
2024 International Journal For Numerical and Analytical Methods in Geomechanics
doi: 10.1002/nag.3857
Cohesive material columns have been extensively used in foundation improvement projects to enhance the bearing capacity of composite foundations and mitigate post‐construction settlement. However, the low permeability of cohesive material columns restricts the dissipation of pore water primarily through the top surface of the foundation, potentially resulting in longer drainage paths compared to foundations treated with granular material columns or vertical drains. Moreover, the impact of non‐Darcian flow within soils on consolidation behavior becomes increasingly pronounced as the drainage path increases. Consequently, a novel analytical model for the consolidation of impervious column‐assisted foundations is established, which can incorporate the seepage model accounting for the initial hydraulic gradient. The accuracy and reasonableness of the obtained solution are then validated by conducting a comparative analysis with existing models and through a detailed case study. Furthermore, a parametric analysis is conducted to delve into the influence of several crucial factors on the consolidation performance. The findings demonstrate that non‐Darcian flow has a greater influence on composite foundations compared to natural foundations. Additionally, the threshold value of the well‐diameter ratio decreases with the increase in the initial hydraulic gradient. Finally, the final seepage front remains at a shallower position when the column–soil modulus ratio becomes larger, and the influence of non‐Darcian flow on the consolidation rate becomes more pronounced.
Efficient Random Field Generation With Rotational Anisotropy for Probabilistic SPH Analysis of Slope FailureBi, Zhonghui; Wu, Wei; Zhang, Liaojun; Peng, Chong
2024 International Journal For Numerical and Analytical Methods in Geomechanics
doi: 10.1002/nag.3858
Due to geological processes such as sedimentation, tectonic movement, and backfilling, natural soil often exhibits characteristics of rotated anisotropy. Recent studies have shown the significant impact of rotated anisotropy on slope stability. However, little research has explored how this rotated anisotropy affects the large deformations occurring after slope failure. Therefore, this study integrates rotated random field theory with smoothed particle hydrodynamics (SPH) to investigate its influence on post‐failure slope behavior. Focusing on a typical slope scenario, this research utilizes graphics processing unit (GPU)–accelerated covariance matrix decomposition (CMD) method to create rotated anisotropy random fields and applies the SPH framework for analysis. It examines the influence of rotated anisotropy angles and the cross‐correlation between cohesion and internal friction angle on landslides. The results indicate that the rotational anisotropy of the slope significantly influences post‐failure behavior. When the rotation angle is close to the slope surface, it tends to amplify both the magnitude and variability of slope failure. Furthermore, the study evaluates the efficiency of generating these random fields and emphasizes the substantial computational speed improvements achieved with GPU acceleration. These findings offer a robust approach for probabilistic analysis of slope large deformations considering rotated anisotropy. They provide a theoretical foundation for accurately assessing the risk of slope collapse, holding significant practical implications for geotechnical engineering.
A Semi‐Analytical Method for Simulating Near‐Field Antiplane Wave Propagation in Layered Fluid‐Saturated Porous MediaLi, Liang; Wang, Man; Jiao, Hongyun; Du, Xiuli; Shi, Peixin
2024 International Journal For Numerical and Analytical Methods in Geomechanics
doi: 10.1002/nag.3859
A semi‐analytical method for the near‐field antiplane wave propagation analysis in the layered fluid‐saturated porous media (FSPM) is proposed based on the Biot u–U dynamic formulation. The wave propagation equations of the FSPM are decoupled by the variable‐separating method. The thin‐layer element method (TLEM) is applied to discretize the infinite domain and construct the consistent artificial boundary condition. The finite element method (FEM) is adopted for the space discretization of the finite domain and the numerical solution of the dynamic response. The proposed method is validated by the comparison of the numerical results of this method with those in the published references and acquired from the remote artificial boundary. Subsequently, this method is applied to investigate typical near‐field antiplane wave propagation problems in the FSPM. Parametric sensitivity investigations are also executed to explore the impact of mechanical parameters, including permeability coefficients, porosity, and shear modulus of the solid phase, on the dynamic response of the FSPM. The study results confirm the efficacy and efficiency of the proposed method in the near‐field antiplane wave propagation analysis in the FSPM.
Implementation of the Zienkiewicz–Pande Model into a Four‐Dimensional Lattice Spring Model for Plasticity and FractureWei, Xin‐Dong; Li, Zhe; Zhao, Gao‐Feng
2024 International Journal For Numerical and Analytical Methods in Geomechanics
doi: 10.1002/nag.3860
Plasticity and fracture problems have always been hot topics in numerical methods. In this work, a universal implementation procedure for the elasto‐plastic constitutive model is developed in the four‐dimensional lattice spring model (4D‐LSM), in which the Jaumann stress rate is incorporated to exclude the influence of the rigid rotation in the particle stress, expanding the ability of 4D‐LSM to deal with large elastic deformation problems by its own to large plastic deformation problems. As an example, the Zienkiewicz–Pande (ZP) constitutive model is implemented. Several numerical examples are carried out to check the performance of the implemented model. Through a comparison with analytical solutions, available experimental data, and other numerical results, the stability of the developed plastic framework and the correctness of the stress calculation scheme are verified. Meanwhile, numerical results show that the developed code is capable of solving elasto‐plastic large deformation problems. With the advantage of 4D‐LSM in handling fracture problems, the ability of the embedded model to solve plastic fracture problems is verified with a simple maximum deformation failure criterion.
Microscopic Thermo‐Mechanical Properties and Phase Transition of Bulk Ice‐IhWei, Pengchang; Niu, Weiwei; Yao, Chi; He, Zhenyu; Zheng, Yuan‐Yuan; Ma, Wei
2024 International Journal For Numerical and Analytical Methods in Geomechanics
doi: 10.1002/nag.3856
The ice–water phase transition of bulk ice could develop with varying temperatures and external loads, significantly affecting its mechanical properties. The coupling effect of temperature and shear loads on the thermo‐mechanical properties of bulk ice and its phase transition evolution is poorly understood, especially at the nanoscale. In this study, molecular dynamics (MD) simulation method was employed to investigate the thermo‐mechanical behaviours of bulk ice‐Ih system at the microscale under various temperatures (73–270 K) and shear paths, where its phase transition, elastic properties, structure deformation mechanism and structural anisotropy were discussed. The simulation results show that (1) the shear modulus, shear strength and ultimate shear strain of bulk ice‐Ih system could linearly decrease with rising temperature, aligning with previous studies. (2) Two types of failure modes from bulk ice‐Ih system were founded, such as solid–liquid phase co‐existence at 73–225 K and liquid phase at 250–270 K. (3) Ice melting into water was attributed to the fracture of hydrogen bond during shear process. (4) Compared to vertical shearing (XZ (112¯0$11\bar{2}0$) and YZ (011¯0$01\bar{1}0$)) directions, the mechanical response along the horizontal shearing (XY (0001)) direction was most sensitive to temperature effect.
Evaluation of the Spatial Variability of the Mechanical Properties of Rocks Using Non‐Iterative Green's Function Approach and the FOSM MethodMesquita, Leonardo C.; Sotelino, Elisa D.; Peres, Matheus L.
2024 International Journal For Numerical and Analytical Methods in Geomechanics
doi: 10.1002/nag.3861
The present work proposes a new version of the Green‐FOSM (first‐order second moment) method, which eliminates the iterative calculation process of the original version and, simultaneously, solves the convergence problems related to the mechanical properties of rocks that form the geological formation. In this calculation scheme, the iterative process is eliminated by using a matrix that correlates the nodal displacement vector with the strain vector. Considering the same computational resources, this non‐iterative version of the Green‐FOSM method is up to 200 times faster than the original iterative process. In addition, it allows analyzing problems with more than 10,000 random variables, value that in the original method is less than 3000. To demonstrate its validity, the proposed method is applied to two hypothetical models subjected to different fluid extraction processes. For all the different levels of correlation and spatial variability, the statistical results obtained by the proposed method agree well with the results obtained via Monte Carlo Simulation (MCS). The relationship between CPU times demonstrates that the proposed method is at least 50 times faster than MCS. In the end, the non‐iterative Green‐FOSM method is used to obtain the displacement, strain, and stress fields of a geological section constructed from a seismic image of Brazilian pre‐salt oil region. The results found show that, depending on the levels of spatial variability, the analyzed fields can assume values up to 30.6% higher or lower than the values obtained deterministically.
Use of Advanced Constitutive Models for the Mechanical Behavior of Soft Soils With Diatoms From Bogotá (Colombia)Mendoza, Cristhian; Farias, Márcio Muniz
2024 International Journal For Numerical and Analytical Methods in Geomechanics
doi: 10.1002/nag.3863
Most constitutive models did not initially consider special behaviors in some soils with singular characteristics (e.g., soft soils with diatom content). For example, at first, these models did not consider the effect of soil structure and viscosity. However, in the last decades, these variables have been incorporated into several constitutive models to describe the mechanical behavior of the soil in its natural state. Structure and viscosity laws that adequately reproduce the soil behavior had to be developed to include these variables. This paper compares the mechanical behavior of soft soils in Bogotá with different constitutive models. Bogotá’s soft soils are lacustrine deposits with a high content of diatoms in their structure. Natural soil samples with intact structures show a high‐water content, which can be higher than 300%, liquid limits of up to 400%, void ratios higher than five, and friction angles of almost 40°. In addition, the model validations were made through the simulations of triaxial tests in compression and shear paths. Modified Cam Clay (MCC), hypoplastic (HP), and subloading Cam Clay (SCC) were the constitutive models used. Two models are based on an elastoplastic framework, and the third uses a HP framework. Several lessons were learned from the simulations regarding the strengths and weaknesses of the models compared to the tests carried out. Finally, the extensive discussion revolves around determining the most suitable model for simulating the mechanical behavior of soft soils containing diatoms in Bogotá.
High‐Fidelity Data Augmentation for Few‐Shot Learning in Jet Grout Injection ApplicationsAtangana Njock, Pierre Guy; Yin, Zhen‐Yu; Zhang, Ning
2024 International Journal For Numerical and Analytical Methods in Geomechanics
doi: 10.1002/nag.3862
Contemporary geoengineering challenges grapple with the plateauing of both existing algorithms and their depth of insights, a phenomenon exacerbated by the scarcity of high‐fidelity data. Although existing solutions such as Monte‐Carlo method can generate abundant data, they are not sufficiently robust for ensuring the high fidelity of data. This study proposes a novel data augmentation framework that combines statistical and machine learning methods to generate high‐fidelity synthetic data, which closely align with field data in terms of the statistical and empirical attributes. The innovations of the proposed approach lie in the integration of Copulas theory for data generation, a developed geo‐regression anomaly detection (GRAD) for adjusting data attributes, and an evolutionary polynomial regression for data consistency enforcement. The multilayer perceptron (MLP) and a wide‐and‐deep (WaD) networks are applied to assess the effectiveness of high‐fidelity data augmentation using jet grouting data. The outcomes reveal the robustness of the synthetic data generation framework, achieving satisfactory fidelity in both empirical and statistical attributes. The proposed data augmentation improved the R2 and MAE achieved by MLP and WaD up to 28.37% under data fractions ranging from 0.2 to 1. MLP and WaD yielded comparable results in terms of accuracy and generalization ability across various augmented fractions. This indicates that the accuracy of synthetic data plays a pivotal role, suggesting improving data quality can be highly effective in boosting performance, regardless of the model complexity. This study contributes valuable insights to addressing the challenges of scare high‐fidelity data in geoengineering.