Comparative study on mesh-based and mesh-less coupled CFD-DEM methods to model particle-laden flow

TL;DR

This paper compares mesh-based (FVM) and mesh-less (SPH) CFD methods coupled with DEM for modeling particle-laden flows, highlighting their similarities, differences, and validation through three test cases.

Contribution

It introduces a new boundary condition model for SPH and provides a detailed comparison with FVM in coupled CFD-DEM simulations.

Findings

DEM-FVM and DEM-SPH show good agreement with analytical results.

Differences mainly due to local fluid fraction calculations.

Mesh-less and mesh-based methods are both viable for particle-laden flow modeling.

Abstract

A comparative study on mesh-based and mesh-less Computational Fluid Dynamics (CFD) approaches coupled with the Discrete Element Method (DEM) is presented. As the mesh-based CFD approach a Finite Volume Method (FVM) is used. A Smoothed Particle Hydrodynamics (SPH) method represents mesh-less CFD. The unresolved fluid model is governed by the locally averaged Navier-Stokes equations. A newly developed model for applying boundary conditions in the SPH is described and validation tests are performed. With the help of the presented comparative tests, the similarities and differences of DEM-FVM and DEM-SPH methods are discussed. Three test cases, comprised of a single solid particle sedimentation test, flow through a porous block and sedimentation of a porous block, are performed using both methods. Drag forces acting on solid particles highly depend on local fluid fractions. For comparative reasons, the size of a cell in FVM is chosen such that fluid fractions match those computed in SPH. In general, DEM-FVM and DEM-SPH methods exhibit good agreement with analytic reference results. Differences between DEM-SPH and DEM-FVM approaches were found mostly due to differences in computed local fluid fractions.