Search

The structured sand presents significant interparticle bonding and anisotropy, resulting in significant differences in the physical and mechanical properties from the pure sand. This study proposes a new surrogate model based on the concept of multi‐fidelity residual neural network (MR‐NN) as an alternative to DEM simulation for predicting the mechanical behaviours of structured sand with different initial anisotropy and saving largely computational costs. The model is initially trained using low‐fidelity (LF) data to focus on capturing the main underpinning correlations between macroscopic mechanical parameters with inter‐particle properties and anisotropic state variables (i.e., microscopic fabric and tilting angle of sand), where the LF data are generated from previously proposed anisotropic macro‐micro quantitative correlation. Subsequent training on sparser high‐fidelity (HF) data is used to calibrate and refine the model with HF data generated from DEM simulations for anisotropic structured sand. Feedforward neural network (FNN) is adopted as the baseline algorithm for training models. The macroscopic anisotropic parameters of structured sand predicted by the surrogate model are compared with DEM simulations to examine its feasibility and generalization ability. Furthermore, the robustness of the surrogate model is examined by discussing the effect of LF data on the performance of MR‐NN. The superiority of the MR‐NN is further verified by comparing the performance of the trained MR‐NN with the one‐shot trained FNN based on the same HF data. All results demonstrate that the proposed surrogate model can provide a fast and accurate simulation of the anisotropic parameters of structured sand. [ABSTRACT FROM AUTHOR]

Using the microstructural mechanics approach, the macroscopic anisotropic failure criteria expressed in closed form by the microscopic parameters are derived for regularly arranged elliptical particle assemblies. The close-packed, elliptical particle assembly is equivalent to a lattice model that is described by a lattice beam system, and the Mohr–Coulomb-like shear failure criterion is applied to describe the beam breakages. Through mechanical analysis of the instability state of the unit cell of the lattice beam system, macroscopic anisotropic strength criteria and the corresponding strength parameters, which are expressed by microscopic parameters and the tilting angle, are proposed. The macroscopic failure characteristics and strength parameters are qualitatively consistent with the results of the discrete element method (DEM), and the influences of the microscopic parameters and tilting angle on the macroscopic strength parameters are investigated in detail according to the proposed theoretical results. The applicability of the theoretical results to randomly distributed particles is explored by DEM simulation of irregularly arranged particles, and the deviation are also systematically discussed. [ABSTRACT FROM AUTHOR]

Using a microstructural mechanics approach, the close-packed elliptical particle assembly is first represented by a lattice model that is described by a beam system, and the Mohr–Coulomb (MC) shear failure criterion is used to describe the contact (beam) breakage. Through structural mechanics analysis of the unit cell, macroscopic strength criteria and associated parameters at two anisotropic principal directions are expressed in closed form with respect to the particle shape and microscopic strength parameters while considering the successive breakage of contacts. The macroscopic failure characteristics of elliptical particle assemblies are described by the MC failure criterion, and the macro–micro relations of strength parameters can be directly determined by the proposed expressions. The proposed analytical expressions are confirmed to be accurate due to the good consistency between theoretical and discrete element method (DEM) results based on the regular arrangement. The introduction of the fabric anisotropy parameter extends the applicability of the theoretical results to irregular arrangement cases, and the deviation with more realistic particle assemblies is investigated. The proposed expressions of the macroscopic strength parameters can help us better understand the influence of inherent anisotropy and microscopic properties and provide an alternative approach to determine the microscopic parameters in DEM simulations. [ABSTRACT FROM AUTHOR]