Feasibility study on coupled CFD-DEM modeling of fluid-driven soil erosion at the particle scale

Abstract

Understanding and modeling of the fluid-particle interaction is important in modeling soil erosion processes. However, computation of particle-level forces applied by fluid flows and thus modeling of erosion behaviors at a particle level have been daunting tasks. Therefore, this study presents a numerical approach to model the fluid-driven erosion of non-cohesive sands by coupling computational fluid dynamics (CFD) and discrete element method (DEM). Here, we modeled the erosion function apparatus (EFA) test which is frequently used to estimate soil erodibility or soil erosion resistance in the laboratory. DEM was used to model the particle motions by solving Newton's laws of motion and CFD was employed to simulate the fluid flows by solving the volume-averaged Navier-Stokes equations. In addition, the k-ω turbulence model was implemented into the fluid phase to describe turbulent flow behaviors. The particle-level forces, including drag force, buoyancy force, pressure-gradient force, and viscous force, were computed and taken into account for the fluid-particle interactions during the fluid-driven erosion process. The simulation results were discussed in comparison to the experimental EFA test results. This work shows that the coupled CFD-DEM can be a useful and promising tool to model and analyze the soil erosion behavior at the particle scale.