Particle characteristics (particle shape and size), along with relative density, significantly influence the frictional characteristics and liquefaction behavior of granular materials, particularly sand. While many studies have examined the individual effects of particle shape, gradation, and relative density on the frictional characteristics and liquefaction behavior of sand, they have often overlooked the combined effects of these soil parameters. In this study, the individual effect of these three soil parameters on the strength characteristics (angle of internal friction) and liquefaction resistance has been quantified by analyzing the data available in the literature. A novel dimensionless parameter, the 'packing index (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha $$\end{document}),' was developed to account for the bulk characteristics (relative density - RD) and grain properties (gradation, represented by the coefficient of uniformity (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$C_u$$\end{document}), and particle shape represented by the shape descriptor regularity (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\rho $$\end{document})) of the granular soils. Through statistical analysis, a power law-based equation was proposed and validated to relate the cyclic resistance ratio (CRR) and angle of internal friction (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\phi $$\end{document}) with the packing index. Finally, an approach to assess the liquefaction resistance was detailed considering the intrinsic soil parameters, aiming to bridge the gap between field observations and laboratory analysis to facilitate a comprehensive understanding of soil behavior under cyclic loading.

The shape of particles significantly influences their mechanical properties, making accurate shape modeling crucial in numerical simulations. This paper proposes a framework for generating particles by applying improved spherical harmonic reconstructions to convex hull surfaces. The framework integrates mesh refinement tech- niques to enhance mesh resolution, enabling the generation of finer surface details than 3D laser scanning. Three parameters are introduced: Delta K1, which controls roundness; Delta K2, which governs roughness; and Rd, which represents the boundary between roundness and roughness in spherical harmonic reconstructions. Introducing these parameters not only allows independent control over the three levels of shape (form, roundness, and roughness) but also enhances the flexibility of the method, enabling the generation of various particle shapes. Granular assemblies with varying roundness and roughness distributions are generated and applied in discrete element method (DEM) simulations of triaxial shear. The results show that roundness is negatively correlated with the peak friction angle, while roughness is positively correlated. The proposed method enhances the ability to generate complex particle shapes, offering a practical tool for modeling and simulating granular materials.

Particle shape is an intrinsic characteristic of soil particles that significantly influences mechanical responses. In this investigation, a meticulously calibrated and validated two-dimensional discrete element method (DEM) model of a biaxial shearing test was employed to simulate the shearing response of forty distinct particle shapes. The systematic evolution of particle roundness (R) and aspect ratio (AR) was achieved by utilizing idealized polygonal-shaped particles, aiming to comprehend their effects on the macro and micromechanical behaviors of granular materials. The results suggest that a reduction in R limits free rotations and enhances interlocking, thereby promoting relatively stable force transmission between particles and leading to a monotonic increase in shear strength. However, this effect diminishes as particles become more elongated. Conversely, a decrease in AR from 1.0 (increased elongation) constrains particle rotations, increases the coordination number, and enhances fabric anisotropy initially resulting in increased overall shear strength, reaching a maximum before exhibiting a decreasing trend, indicative of non-monotonic variation. For high elongations, notable fabric anisotropy impedes clear force transmission between particles thus facilitating interparticle sliding and overall strength diminishes. The extent to which AR impacts depends on the angularity feature of particles. Finally, a nonlinear equation has been proposed to predict the variation in critical state shear strength of granular samples, based on the R and AR values of the constituent particles.

The deformation characteristics and constitutive behavior of granular materials under normal forces acting on particles are dependent on the geometry of the grain structure, fabrics and the inter-particle friction. In this study, the influence of particle morphology on the friction and dilatancy of five natural sands was investigated using deep learning (DL) techniques. A Three-dimensional (3D) imaging technique using computed tomography was utilized to compute the morphology (roundness and sphericity) of collected natural sands. Triaxial tests were conducted on the five different natural sands at different densities and confinement stresses (r3). From the triaxial results, peak friction angle (up), critical state friction angle (ucs), and dilatancy angle (w) were obtained and modeled using conventional multiple linear regression (MLR) models and DL techniques. A total of 100 deep artificial neural networks (DANN) models were trained at different sizes of first and second hidden layers. The use of MLR resulted in R2 of 0.709, 0.565, and 0.795 for up, u cs and w, respectively, while the best performed DANN (30 and 50 neurons for the 1st and 2nd hidden layers, respectively) had R2 of 0.956 for all outputs (up, u cs and w) combined. Using the best-performed DANN model, the weight partitioning technique was used to compute an importance score for each parameter in predicting up, u cs and w. The r3 had the highest importance followed by relative density, roundness, and sphericity with a relative importance of more than 10%. In addition, sensitivity analysis was conducted to investigate the effect of each parameter on the shear parameters and ensure the robustness of the developed model. (c) 2024 Production and hosting by Elsevier B.V. on behalf of The Japanese Geotechnical Society. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

The mechanical properties of sandy soil, a prevalent material in engineering construction, are affected by several factors, including the particle shape. This study utilized a rigid block unit for particle simulation and established specimens reflecting the real particle appearance based on mu CT scanning geometry. Four simple particle types with varying roundness were designed according to the roundness definition, and corresponding numerical test models were developed. These models were employed to examine the effect of particle roundness on the macroscopic mechanical properties of sandy soil using triaxial shearing simulation tests, disregarding other shape parameters. The results demonstrate that particle roundness significantly affects the mechanical properties of sandy soil. Within a roundness range of [0,1], samples with varying particle roundness displayed different deviatoric stress-strain curves. As the roundness increases, the peak deviatoric stress decreases, and both the cohesive strength and friction angle of the samples reduce.