Search results

This paper aims to present a novel scheme for imposing periodic boundary conditions with downscaled macroscopic strain measures of gradient Cosserat continuum on the representative volume element (RVE) of discrete particle assembly in the frame of the second-order computational homogenization methods for granular materials.

The proposed scheme is based on the generalized Hill’s lemma of gradient Cosserat continuum and the incremental non-linear constitutive relation condensed to the peripheral particles of the RVE of discrete particle assembly. The generalized Hill’s lemma conducts to downscale the macroscopic strain or stress measures and to impose the periodic boundary conditions on the RVE boundary so that the Hill-Mandel energy equivalence condition is ensured. Because of the incremental non-linear constitutive relation condensed to the peripheral particles of the RVE, the periodic boundary displacement and traction constraints together with the downscaled macroscopic strains and strain gradients, micro-rotations and curvatures are imposed in the point-wise sense without the need of introducing the Lagrange multipliers for enforcing the periodic boundary displacement and traction constraints in a weak sense.

Numerical results demonstrate that the applicability and effectiveness of the proposed scheme in imposing the periodic boundary conditions on the RVE. The results of the RVE subjected to the periodic boundary conditions together with the displacement boundary conditions in the second-order computational homogenization for granular materials provide the desired estimations, which lie between the upper and the lower bounds provided by the displacement and the traction boundary conditions imposed on the RVE respectively.

Each grain in the particulate system under consideration is assumed to be rigid and circular.

The proposed scheme for imposing periodic boundary conditions on the RVE can be adopted solely for estimating the effective mechanical properties of granular materials and/or integrated into the frame of the second-order computational homogenization method with a nested finite element method-discrete element method solution procedure for granular materials. It will tend to provide, at least theoretically, more reasonable results for effective material properties and solutions of a macroscopic boundary value problem simulated by the computational homogenization method.

This paper presents a novel scheme for imposing periodic boundary conditions with downscaled macroscopic strain measures of gradient Cosserat continuum on the RVE of discrete particle assembly for granular materials without need of introducing Lagrange multipliers for enforcing periodic boundary conditions in a weak (integration) sense.

This paper aims to construct smooth poly-ellipsoid shapes from synchrotron microcomputed tomography (SMT) images on sand and to develop a new discrete element method (DEM) contact detection algorithm.

Voxelated images generated by SMT on Colorado Mason sand are processed to construct smooth poly-ellipsoidal particle approximations. For DEM contact detection, cuboidal shape approximations to the poly-ellipsoids are used to speed up contact detection.

The poly-ellipsoid particle shape approximation to Colorado Mason sand grains is better than a simpler ellipsoidal approximation. The new DEM contact algorithm leads to significant speedup and accuracy is maintained.

The paper limits particle shape approximation to smooth poly-ellipsoids.

Poly-ellipsoids provide asymmetry of particle shapes as compared to ellipsoids, thus allowing closer representation of real sand grain shapes that may be angular and unsymmetric. When incorporated in a DEM for computation, the poly-ellipsoids allow better representation of particle rolling, sliding and interlocking phenomena.

Method to construct poly-ellipsoid particle shapes from SMT data on real sands and computationally efficient DEM contact detection algorithm for poly-ellipsoids.

The purpose of this paper is to determine the representative volume element (RVE) size for quasi‐brittle materials using a discrete approach, namely a lattice‐particle model.

Different material samples are generated and subjected to study regarding its size, maximum aggregate size and boundary conditions. In order to determine the mechanical properties such as the elastic modulus, Poisson's ratio or tensile strength, several tension tests are carried out. For this purpose, a lattice‐particle approach is used to model concrete's fracturing behavior. The information provided by the previous simulations is implemented in a statistical analysis to determine the size of the RVE.

The determination of the RVE size for quasi‐brittle materials is successfully achieved by means of a lattice‐particle model. Computed results show a good agreement with other results reported in the bibliography.

Within a general multiscale framework, the determination of the RVE size is of great interest and some studies have been performed for random heterogeneous materials. However, these analyses are mainly continuum‐based. The estimation of the RVE size is important for correctly predicting the mechanical properties and can be used in different multiscale schemes.

The purpose of this paper is to extend the bridge scale method (BSM) developed for granular materials with only the solid phase to that taking into account the effects of wetting process in porous continuum. The granular material is modeled as partially saturated porous Cosserat continuum and discrete particle assembly in the coarse and fine scales, respectively.

Based on the mass and momentum conservation laws for the three phases, i.e. the solid skeleton, the pore water and the pore air, the governing equations for the unsaturated porous Biot-Cosserat continuum model in the coarse scale are derived. In light of the passive air pressure assumption, a reduced finite element model for the model is proposed. According to the decoupling of the fine and coarse scale calculations in the BSM, the unsaturated porous Cosserat continuum model using the finite element method and the discrete element model using the discrete element method for granular media are combined.

The numerical results for a 2D example problem of slope stability subjected to increasing rainfall along with mechanical loading demonstrate the applicability and performance of the present BSM. The microscopic mechanisms of macroscopic shear band developed in the slope are demonstrated.

Do not account for yet the effects of unsaturated pore water in the fine scale.

The novel BSM that couples the Biot-Cosserat porous continuum modeling and the discrete particle assembly modeling in both coarse and fine scales, respectively, is proposed to provide a micro-macro discrete-continuum two-scale modeling approach for numerical simulations of the hydro-mechanical coupling problems in unsaturated granular materials.

The purpose of this paper is to present knowledge in estimating yield surfaces of heterogeneous media by use of homogenization, especially where the macroscopic behaviour is driven by weak interfaces between phase constituents.

A computational homogenization procedure is used to determine the yield surface of a Representative Volume Element (RVE) that contains a fully debonded inclusion embedded within ideally plastic matrix, whereby the interface is modelled by a Coulomb type friction law.

The macroscopic behaviour of the RVE is shown to coincide an RVE with a hole for expanding loads, whereas for compressive loads, it was shown to approach an RVE with a fully bonded inclusion.

The present paper builds on Gurson’s work in estimating macroscopic yield surfaces of heterogeneous materials. The work is novel in the sense that there had been no previous publications discussing influence of weak interfaces on yield surfaces.

Fibrous porous media have a wide variety of applications in insulation, filtration, acoustics, sensing, and actuation. To design such materials, computational modeling methods are needed to engineer the properties systematically. There is a lack of efficient approaches to build and modify those complex structures in computers. The paper aims to discuss these issues.

In this paper, the authors generalize a previously developed periodic surface (PS) model so that the detailed shapes of fibers in porous media can be modeled. Because of its periodic and implicit nature, the generalized PS model is able to efficiently construct the three-dimensional representative volume element (RVE) of randomly distributed fibers. A physics-based empirical force field method is also developed to model the fiber bending and deformation.

Integrated with computational fluid dynamics (CFD) analysis tools, the proposed approach enables simulation-based design of fibrous porous media.

In the future, the authors will investigate robust approaches to export meshes of PS models directly to CFD simulation tools and develop geometric modeling methods for composite materials that include both fibers and resin.

The proposed geometric modeling method with implicit surfaces to represent fibers is unique in its capability of modeling bent and deformed fibers in a RVE and supporting design parameter-based modification for global configuration change for the purpose of macroscopic transport property analysis.

Granular materials possess multiscale structures, i.e. micro-scales involving atoms and molecules in a solid particle, meso-scales involving individual particles and their correlated structure, and macroscopic assembly. Strong and abundant dissipations are exhibited due to mesoscopic unsteady motion of individual grains, and evolution of underlying structures (e.g. force chains, vortex, etc.), which defines the key differences between granular materials and ordinary objects. The purpose of this paper is to introduce the major studies have been conducted in recent two decades.

The main properties at individual scale are introduced, including the coordination number, pair-correlation function, force and mean stress distribution functions, and the dynamic correlation function. The relationship between meso- and macro-scales is analyzed, such as between contact force and stress, the elastic modulus, and bulk friction in granular flows. At macroscales, conventional engineering models (i.e. elasto-plastic and hypo-plastic ones) are introduced. In particular, the so-called granular hydrodynamics theory, derived from thermodynamics principles, is explained.

On the basis of recent study the authors conducted, the multiscales (both spatial and temporal) in granular materials are first explained, and a multiscale framework is presented for the mechanics of granular materials.

It would provide a paramount view on the multiscale studies of granular materials.

The granular structure of a grinding wheel determines its performance remarkably when grinding. Nowadays, grinding wheels are man-made porous conglomerates of hard abrasive grains bonded with a binder. As an engineered material, it would be interesting to foresee the behaviour of the wheel, but it is hindered under its complex heterogeneous nature. Recently, some models based on the discrete element method (DEM) have been presented to model the grinding wheel. This paper aims to identify and propose a framework that must comply with such models and to identify new applications for DEM models in grinding.

First, the characteristics of the grinding wheel are described. In this way, the framework of DEM models of a grinding wheel material is developed, reproducing both the granular morphology and stiffness.

The paper asserts a promising future that provides the DEM in the modelling of grinding.

The potential of DEM in grinding is analysed, proposing new applications. It can be used as topography model, which can also model the stiffness. In addition, DEM opens a new interesting research line: the modelling of the grit breakout. It draws up the development of essential dressing and wear models.

The purpose of this paper is to further develop the truss‐like discrete element method (DEM) in order to make it suitable to deal with damage and fracture problems.

Finite and boundary elements are the best developed methods in the field of numerical fracture and damage mechanics. However, these methods are based on a continuum approach, and thus, the modelling of crack nucleation and propagation could be sometimes a cumbersome task. Besides, discrete methods possess the natural ability to introduce discontinuities in a very direct and intuitive way by simply breaking the link between their discrete components. Within this context, the present work extends the capabilities of a truss‐like DEM via the introduction of three novel features: a tri‐linear elasto‐plastic constitutive law; a methodology for crack discretization and the computation of stress intensity factors; and a methodology for the computation of the stress field components from the unixial discrete‐element results.

Obtained results show the suitability and the performance of the proposed methodologies to solve static and dynamic crack problems (including crack propagation) in brittle and elasto‐plastic materials. Computed results are in good agreement with experimental and numerical results reported in the bibliography.

This paper demonstrates the versatility of the truss‐like DEM to deal with damage mechanics problems. The approach used in this work can be extended to the implementation of time‐dependent damage mechanisms. Besides, the capabilities of the discrete approach could be exploited by coupling the truss‐like DEM to finite and boundary element methods. Coupling strategies would allow using the DEM to model the regions of the problem where crack nucleation and propagation occurs, while finite or boundary elements are used to model the undamaged regions.

The scope of the truss‐like DEM has been extended. New procedures have been introduced to deal with elastoplastic‐crack problems and to improve the post processing of the stress results.

The purpose of this paper is to obtain a permeability law of a gas flow through a permeable medium using particle image velocimetry experimental data as primal information, which is conflated with numerical calculations by means of a multi-scale method.

The D2Q9 single-relaxation-time Lattice Boltzmann model (LBM) implemented in GPU is used for the numerical calculations. In a first homogenized micro-scale, the drag forces are emulated by means of an effective Darcy law acting only in the close neighborhood of the solid structures. A second mesoscopic level of homogenization makes use of the effective drag forces resulting from the first-scale model.

The procedure is applied to an experiment consisting of a regular array of wires. For the first level of homogenization, an effective drag law of the individual elemental obstacles is produced by conflating particle image velocimetry measurements of the flow field around the wires and numerical calculations performed with a GPU implementation of the LBM. In the second homogenization, a Darcy–Forchheimer correlation is produced, which is used in a final homogenized LBM model.

The numerical simulations at the first level of homogenization require a substantial amount of calculations, which in the present case were performed by means of the computational power of a GPU.

The homogenization procedure can be extended to other permeable structures. The micro-scale-level model retrieves the fluid-structure forces between the flow and the obstacles, which are difficult to obtain experimentally either from direct measurement or by indirect assessment from velocity measurements.