Representative elementary volume analysis of E-B constitutive model parameters of rockfill materials using DEM

Abstract

The engineering application of granular materials is based on continuum mechanics formulated at the macroscopic level. Triaxial compression tests are usually used to determine the equivalent mechanical properties of rockfill materials. However, rockfill materials have a wider range of gradation and the particles are more easily crushed, which is not observed for finely graded materials (e.g. sand, grain), except under high stress levels. So both particle arrangement and particle crushing must be considered for rockfill materials, which makes it more complicated to study the effect of specimen size on rockfill materials and find the representative elementary volume (REV). To investigate this topic, the effect of specimen size on the mechanical behavior of basalt rockfill materials was studied by DEM simulation. First, a series of laboratory single-particle crushing tests were conducted to analyze the crushing characteristics of basalt particles and to determine the particle crushing strength parameters required for DEM simulations. Then the REV of the E-B constitutive model parameters of the rockfill materials is assessed by a series of triaxial compression simulation tests with different specimen sizes, particle arrangements and confining pressures. The results show that for basalt rockfill materials with easily broken particles and a wide range of gradation, particle arrangement and specimen size both impact mechanical behavior. The mean value and standard deviation of each mechanical parameter decrease as specimen size increases. The sizes of the REV with different mechanical parameters are not all the same, and the REV for the E-B constitutive model parameters is 360 × 720 mm (diameter D × height H). It is suggested that the triaxial specimen size should not be less than 6 times the maximum particle size when assessing the E-B constitutive model parameters of the rockfill materials.