Effective simulation of flexible lateral boundaries in two-and three-dimensional DEM simulations

Discrete element method (DEM) models to simulate laboratory element tests play an important role in advancing our understanding of the mechanics of granular material response, including bonded or cemented, particulate materials.Comparisons of the macro-scale response observed in a real physical test and a "virtual" DEM-simulated test can calibrate or validate DEM models.The detailed, particle scale information provided in the DEM simulation can then be used to develop our understanding of the material behaviour.It is important to accurately model the physical test boundary conditions in these DEM simulations.This paper specifically considers triaxial tests as these tests are commonly used in soil mechanics.In a triaxial test,the test specimen of granular material is enclosed within a flexible latex membrane that allows the material to deform freely during testing, while maintaining a specified stress condition. Triaxial tests can only be realistically simulated in 3D DEM codes, however analogue,2D, biaxial DEM simulations are also often considered as it is easier to visualize particle interactions in two dimensions. This paper describes algorithms to simulate the lateral boundary conditions imposed by the latex membrane used in physical triaxial tests in both 2D and 3D DEM simulations.The importance of carefully considering the lateral boundary conditions in DEM simulations is illustrated by considering a 2D biaxial test on a specimen of frictional unbonded disks and a 3D triaxial test on a bonded (cemented) specimen of spheres. The comparisons indicate that the lateral boundary conditions have a more significant influence on the local,particle-scale response in comparison with the overall macro-scale observations.     

Segregation of particulate solids: Experiments and DEM simulations

Segregation of granular materials is a complex phenomenon, difficult to measure quantitatively and to predict. Discrete element method (DEM) can be a useful tool to predict segregation effects and to support the industrial design. In this context, a very challenging idea is the characterization of the granular solids to provide the key parameters needed for a successful DEM simulation of segregation processes. Rolling friction, sliding friction and the coefficient of restitution are the critical parameters to be studied. These microscopic simulation parameters are calibrated by comparing the macroscopic behavior of granular matter in standard bulk experiments, which have the advantage of being highly repeatable and reliable.An experimental method is presented to characterize free surface segregation. The effects of different particle properties, particularly, shape and size, on segregation of cohesionless materials were investigated. From the experiments, particle size demonstrated a stronger effect on segregation than particle shape. Finally, the corresponding DEM simulations of the segregation experiments were presented. The parameters obtained by calibration were validated by the comparison of the modeled segregation behavior with the experimental results. Thus, calibrated DEM simulations are capable of predicting segregation effects.     

Comparison of roller-spreading and blade-spreading processes in powder-bed additive manufacturing by DEM simulations

The roller-spreading and blade-spreading are main powder spreading methods in powder-bed additive manufacturing. The discrete element method was introduced to simulate nylon powder spreading by both roller and blade spreaders. The two spreading processes were compared from several aspects including particle flow behavior, particle contact forces, forces exerted on spreaders, particle segregation and powder layer density. It is found that powder spreading methods mainly affect the movement trajectory of particles, particle contact forces and forces exerted on spreaders. Complicated dispersion and circulation movement of particles occur inside the powder pile by roller-spreading, while particles have relatively weak dispersion by the blade-spreading. The normal force applied to the roller introduces a compacting effect on the powder pile and creates strong force chains that distribute uniformly in the powder pile. Therefore, the powder bed with higher density can be obtained by roller-spreading in thicker powder layer due to the compacting effect. The blade spreader sustains tangential force mainly, so the blade-spreading process limits its application to thicker powder layer. As the powder layer thickness increases, the roller-spreading is more sensitive to segregation index than that of the blade-spreading. The comprehensive comparison of two spreading processes provides criteria for selecting spreading methods.     

DEM simulation of particle percolation in a packed bed

The phenomenon of spontaneous particle percolation under gravity is investigated by means of the discrete element method. Percolation behaviors such as percolation velocity, residence time distribution and radial dispersion are examined under various conditions. It is shown that the vertical velocity of a percolating particle moving down through a packing of larger particles decreases with increasing the restitution coefficient between particles and diameter ratio of the percolating to packing particles. With the increase of the restitution coefficient, the residence time and radial dispersion of the percolating particles increase. The packing height affects the residence time and radial dispersion. But, the effect can be eliminated in the analysis of the residence time and radial dispersion when they are normalized by the average residence time and the product of the packing height and packing particle diameter, respectively. In addition, the percolation velocity is shown to be related to the vertical acceleration of the percolating particle when an extra constant vertical force is applied. Increasing the feeding rate of percolating particles decreases the dispersion coefficient.     

Investigation into improving the efficiency and accuracy of CFD/DEM simulations

The Euler-Lagrange approach combined with a discrete element method has frequently been applied to elucidate the hydrodynamic behavior of dense fluid-solid flows in fluidized beds. In this work, the efficiency and accuracy of this model are investigated. Parameter studies are performed; in these studies, the stiffness coefficient, the fluid time step and the processor number are varied under conditions with different numbers of particles and different particle diameters. The obtained results are compared with measurements to derive the optimum parameters for CFD/DEM simulations. The results suggest that the application of higher stiffness coefficients slightly improves the simulation accuracy. However, the average computing time increases exponentially. At larger fluid time steps, the results show that the average computation time is independent of the applied fluid time step whereas the simulation accuracy decreases greatly with increasing the fluid time step. The use of smaller time steps leads to negligible improvements in the simulation accuracy but results in an exponential rise in the average computing time. The parallelization accelerates the DEM simulations if the critical number for the domain decomposition is not reached. Above this number, the performance is no longer proportional to the number of processors. The critical number for the domain decomposition depends on the number of particles. An increase in solid contents results in a shift of the critical decomposition number to higher numbers of CPUs.     

DEM simulation of particle percolation in a packed bed

The phenomenon of spontaneous particle percolation under gravity is investigated by means of the discrete element method. Percolation behaviors such as percolation velocity,residence time distribution and radial dispersion are examined under various conditions. It is shown that the vertical velocity of a percolating particle moving down through a packing of larger particles decreases with increasing the restitution coefficient between particles and diameter ratio of the percolating to packing particles. With the increase of the restitution coefficient,the residence time and radial dispersion of the percolating particles increase. The packing height affects the residence time and radial dispersion. But,the effect can be eliminated in the analysis of the residence time and radial dispersion when they are normalized by the average residence time and the product of the packing height and packing particle diameter,respectively.In addition,the percolation velocity is shown to be related to the vertical acceleration of the percolating particle when an extra constant vertical force is applied. Increasing the feeding rate of percolating particles decreases the dispersion coefficient.     

Moment-based boundary conditions for straight on-grid boundaries in three-dimensional lattice Boltzmann simulations

In this article, moment-based boundary conditions for the lattice Boltzmann method are extended to three dimensions. Boundary conditions for velocity and pressure are explicitly derived for straight on-grid boundaries for the D3Q19 lattice. The method is compared against the bounce-back scheme using both single and two relaxation time collision schemes. The method is verified using classical benchmark test cases. The results show very good agreement with the data found in the literature. It is confirmed from the results that the derived moment-based boundary scheme is of second-order accuracy in grid spacing and does not produce numerical slip, and therefore offers a transparent way of accurately prescribing velocity and pressure boundaries that are aligned with grid points in three-dimensional.     

Circulation intensity and axial dispersion of non-cohesive solid particles in a V-blender via DEM simulation

In this study, discrete element method （DEM） was employed to simulate the movement of non-cohesive mono-dispersed particles in a V-blender along with particle-particle and particle-boundary interactions. To validate the model, DEM results were successfully compared to positron emission particle tracking （PEPT） data reported in literature. The validated model was then utilized to explore the effects of rotational speed and fill level on circulation intensity and axial dispersion coefficient of non-cohesive particles in the V-blender. The results showed that the circulation intensity increased with an increase in the rotational speed from 15 to 60 rpm. As the fill level increased from 20% to 46%, the circulation intensity decreased, reached its minimum value at a fill level of 34% for all rotational speeds, and did not change significantly at fill levels greater than 34%. The DEM results also revealed that the axial dispersion coefficient of particles in the V-blender was a linear function of the rotational speed. These trends were in good agreement with the experimentallv determined values reported bv previous researchers.     

DEM simulation of polydisperse systems of particles in a fluidized bed

Numerical simulations based on three-dimensional discrete element model (DEM) are conducted for mono-disperse, binary and ternary systems of particles in a fluidized bed. Fluid drag force acting on each particle depending on its size and relative velocity is assigned. The drag coefficient corresponding to Ergun’s correlation is applied to the system of fluidized bed with particle size ratios of 1:1 for the mono-disperse system, 1:1.2, 1:1.4 and 1:2 for the binary system and 1:1.33:2 for the ternary system b...     

Interaction between dry granular materials and an inclined plate (comparison between large-scale DEM simulation and three-dimensional wedge model)

The interaction between dry granular materials and an inclined plate is numerically studied using a three-dimensional discrete element method (DEM) simulation. In the simulation, a plate is dragged horizontally through densely packed dry granular materials. To examine the effect of the rake angle α of the plate on the drag force acting on the plate, three cases with α = 50°, 70°, and 90° are compared (α = 90° for a vertical plate). The results show that for all cases, the force oscillates as the plate advances. As α decreases, the amplitude and frequency of the force oscillation decrease and increase, respectively. The force oscillation is attributed to the periodic evolution of a shear band formed in the materials. The relationships between the rake angle, evolution of the shear band, and drag force can be explained quantitatively by using a three-dimensional wedge model considering the variation of the local volume fraction inside the shear band.     

A formal finite element approach for open boundaries in transport and diffusion ground-water problems

In the application of the finite element method to diffusion and convection-dispersion equations over a ground-water domain, the Galerkin technique was used to incorporate Neumann (or second-type) and Cauchy (or third-type) boundary conditions. While mass movement through open boundaries is a priori unknown, these boundaries are usually treated as a zero Neumann condition at some far distance from the domain of interest. Nevertheless, cheaper and better solutions can be obtained if these unknown conditions are adequately incorporated in the weak formulation and in the transient solution schemes (open boundary condition). Theoretical and numerical proofs are given of the equivalences between this approach and a ‘well-posed’ problem in a semi-infinite domain with a zero Neumann condition at a boundary placed at infinity. Transport and diffusion equations were applied in one dimension to show the numerical performances and limitations of this procedure for some linear and non-linear problems. No a priori limitations are foreseen in order to find similar solutions in two or three dimensions. Thus the spatial discretization in the proximity of open boundaries could be drastically reduced to the domain of interest.     

Equivalent boundary model of lunar soil drilling simulation by DEM

Aiming to solve the computational cost problem in the discrete element simulation for lunar soil drilling sampling, an equivalent boundary method was proposed. A high-accuracy DEM model of lunar soil was established firstly. As the novel alterable constitutive law, the accuracy of the model was verified to meet the performance of real lunar soil very much both in shear strength indices and elastic–plastic behavior. A common drill bit in the geological exploration field for sampling soil was chosen as the simulation object. In preanalysis, it was known that with the increase of drilling depth, the stress concentration area was always near the drill bit, while the affected area of the lunar soil was a cylindrical area around the drill pipe, which extended towards the drilling direction instead of extending around it. Then a big boundary drilling simulation scene was established to investigate the flow direction of lunar soil particles. The motion law of particles and the velocity field information were obtained, and a U-shape chain was described around the drill bit. Finally an equivalent boundary was set near the U-shaped chain, and the size was determined by comparing the soil stress in the fierce collision zone and around the reference boundary. This method could be a reference for other lunar soil drilling researches with other drills of different sizes.     

Application of DEM modified with enlarged particle model to simulation of bead motion in a bead mill

We applied the discrete element method （DEM） of simulation modified by an enlarged particle model to simulate bead motion in a large bead mill. The stainless-steel bead mill has inner diameter of 102 mm and mill length of 198 mm. The bead diameter and filling ratio were fixed respectively at 0.5 mm and 85%. The agitator rotational speed was changed from 1863 to 3261 rpm. The bead motion was monitored experimentally using a high-speed video camera through a transparent mill body. For the simulation, enlarged particle sizes were set as 3-6 mm in diameter. With the DEM modified by the enlarged particle model, the motion of enlarged particles in a mill was simulated.The velocity data of the simulated enlarged particles were compared with those obtained in the experiment. The simulated velocity of the enlarged particles depends on the virtual frictional coefficient in the DEM model. The optimized value of the virtual frictional coefficient can be determined by considering the accumulated mean value. Results show that the velocity of the enlarged particles simulated increases with an increase in the optimum virtual frictional coefficient, but the simulated velocity agrees well with that determined experimentally by optimizing the virtual frictional coefficient in the simulation. The computing time in the simulation decreases with increased particle size.     

Application of the FEM/DEM and alternately moving road method to the simulation of tire-sand interactions

The three-dimensional finite-discrete element method (FEM/DEM) is applied to the simulation of tire-sand interactions, where the tire is discretized into hexahedron finite elements and sand is modeled by using the discrete element method. The feasibility and effectiveness of the method are proven by comparing the simulation results with the current reported results. Since long test roads are usually required for investigating tire running behaviors, which lead to large-scale simulation models and time consuming problems, the alternately moving road method is proposed to handle this problem. It can simulate tire running behaviors on an arbitrary length sand road with a constant road length value. The numerical model of a lug tire running on a bisectional road with fine and coarse sand is established to verify the feasibility of the method.     

Euler-Lagrange/DEM simulation of wood gasification in a bubbling fluidized bed reactor

We present an Euler–Lagrange method for the simulation of wood gasification in a bubbling fluidized bed. The gas phase is modeled as a continuum using the 2D Navier–Stokes equations and the solid phase is modeled by a Discrete Element Method(DEM)using a soft-sphere approach for the particle collision dynamic. Turbulence is included via a Large-Eddy approach using the Smagorinsky sub-grid model.The model takes into account detailed gas phase chemistry,zero-dimensional modeling of the pyrolysis and gasification of each individual particle,particle shrinkage,and heat and mass transfer between the gas phase and the particulate phase.We investigate the influence of wood feeding rate and compare exhaust gas compositions and temperature results obtained with the model against experimental data of a laboratory scale bubbling fluidized bed reactor.     

Application of DEM modified with enlarged particle model to simulation of bead motion in a bead mill

We applied the discrete element method (DEM) of simulation modified by an enlarged particle model to simulate bead motion in a large bead mill. The stainless-steel bead mill has inner diameter of 102 mm and mill length of 198 mm. The bead diameter and filling ratio were fixed respectively at 0.5 mm and 85%. The agitator rotational speed was changed from 1863 to 3261 rpm. The bead motion was monitored experimentally using a high-speed video camera through a transparent mill body. For the simulation, enlarged particle sizes were set as 3–6 mm in diameter. With the DEM modified by the enlarged particle model, the motion of enlarged particles in a mill was simulated. The velocity data of the simulated enlarged particles were compared with those obtained in the experiment. The simulated velocity of the enlarged particles depends on the virtual frictional coefficient in the DEM model. The optimized value of the virtual frictional coefficient can be determined by considering the accumulated mean value. Results show that the velocity of the enlarged particles simulated increases with an increase in the optimum virtual frictional coefficient, but the simulated velocity agrees well with that determined experimentally by optimizing the virtual frictional coefficient in the simulation. The computing time in the simulation decreases with increased particle size.     

Discrete element method simulations of Mars Exploration Rover wheel performance

Mars Exploration Rovers (MERs) experienced mobility problems during traverses. Three-dimensional discrete element method (DEM) simulations of MER wheel mobility tests for wheel slips of i = 0, 0.1, 0.3, 0.5, 0.7, 0.9, and 0.99 were done to examine high wheel slip mobility to improve the ARTEMIS MER traverse planning tool. Simulations of wheel drawbar pull and sinkage MIT data for i ⩽ 0.5 were used to determine DEM particle packing density (0.62) and contact friction (0.8) to represent the simulant used in mobility tests. The DEM simulations are in good agreement with MIT data for i = 0.5 and 0.7, with reasonable but less agreement at lower wheel slip. Three mobility stages include low slip (i < 0.3) controlled by soil strength, intermediate slip (i ∼ 0.3–0.6) controlled by residual soil strength, and high slip (i > 0.6) controlled by residual soil strength and wheel sinkage depth. Equilibrium sinkage occurred for i < 0.9, but continuously increased for i = 0.99. Improved DEM simulation accuracy of low-slip mobility can be achieved using polyhedral particles, rather than tri-sphere particles, to represent soil. The DEM simulations of MER wheel mobility can improve ARTEMIS accuracy.     

Oblique water entry of a cone by a fully three-dimensional nonlinear method

The hydrodynamic problem of a cone entering the water surface obliquely has been analyzed by the three-dimensional (3-D) incompressible velocity potential theory with the fully nonlinear boundary conditions on the moving free surface and body surface boundary. The time stepping method is used in the stretched coordinate system defined as the ratio of the physical system to the distance that the cone has travelled into water. The boundary element method is used to solve the potential at each time step. Both triangular element mesh and quadrilateral element mesh have been used. Discretisation of the body surface and the free surface is applied regularly during the simulation to account for their change and deformation, and data from the old mesh is transferred into the new one through interpolation. Both the dynamic and kinematic free surface boundary conditions are satisfied through the Eulerian form. In particular the free surface elevation and potential variation are traced at a given azimuth of the cylindrical coordinate system, in the direction parallel to the body surface or perpendicular to the free surface to avoid multi-valued function. Detailed convergence study with respect to time step and element size has been undertaken and high accuracy has been achieved. Results for the cone in vertical entry are compared with those obtained from the 2-D axisymmetric method and good agreement is found. Simulations are made for cones of various deadrise angles and different oblique entries and detailed results are provided.     

Boundary integral equations for 2D elasticity and its application in discrete dislocation simulation in finite body: 2. Numerical implementation

In the previous paper by Yu and Diab (2013), several sets of boundary integral equations are derived for general anisotropic materials and corresponding equations for materials with different classes of symmetry are deduced. The work presented herein implements two sets of boundary element schemes to numerically solve the stress field. The integration on the element that has the singular point of the kernel is bounded and can be evaluated analytically. Four benchmark elastic problems are solved numerically to show the advantage of the two schemes over the conventional boundary element formulation in eliminating the boundary layer effect. The one with the weaker singularity has better convergence and gives more accurate results. The presented formulation also provides a direct approach to solve for stress field in a finite solid body in the presence of dislocations. Combined with discrete dislocations dynamics, boundary value problems with dislocations in finite bodies can be solved. Two examples, bending of a single crystal beam and pure shearing of a polycrystalline solid, are simulated by discrete dislocation dynamics using the scheme that has the weaker singularity. The comparisons with the published results using the well-established superposition technique validate the proposed formulation and show its quick convergence.