Spatial and temporal coarse-graining for DEM analysis

Abstract

The discrete element method is used extensively for the simulation of granular materials. The great majority of engineering applications require bulk design parameters to be determined which requires some averaging or coarse graining techniques to be applied to the particle data from DEM computations. Although numerous papers have been published on these techniques, we suggest that there are still significant challenges in the calculation of the continuum quantities from the particle data. This is chiefly because DEM computes at grain contact level and at the order of 10−5 or smaller computational time step, whereas important engineering events often occur at much larger timescale, so the question of what temporal and spatial averaging scales should be adopted is not so clear. This paper outlines an implementation of coarse graining methods for the projection of discrete quantities into continuum fields. Two DEM simulation examples with a quasi-static and a rapid dynamic flow conditions are used to study the influence of the averaging length scales. The results show that the combined temporal and spatial regimes can provide an effective method to evaluate the salient continuum quantities at the appropriate frequency of occurrence. They also show that the computed continuum parameters can be very sensitive to the averaging length scales such that erroneous conclusions can be reached depending on whether an appropriate scale has been chosen.