A Novel CFD-DEM Coarse-Graining Method Based on the Voronoi Tessellation

TL;DR

This paper introduces a new two-grid coarse-graining method using Voronoi tessellation to improve local porosity estimation in CFD-DEM simulations, enhancing force prediction accuracy for polydisperse particles.

Contribution

The paper presents a novel Voronoi tessellation-based coarse-graining approach that captures local porosity effects more accurately than existing methods in CFD-DEM simulations.

Findings

Conserves porosity data effectively.

Provides reasonably grid-independent results.

Improves prediction of fluid-particle forces in polydisperse systems.

Abstract

In unresolved flow CFD-DEM simulations, the porosity values for each CFD cell are determined using a coarse-graining algorithm. While this approach enables coupled simulations of representative numbers of particles, the influence of the porosity local to the particles on the fluid-particle interaction force is not captured. This contribution considers a two-grid coarse-graining method that determines a local porosity for each particle using a radical Voronoi tessellation of the system. A relatively fine, regular point cloud is used to map these local porosity data to the CFD cells. The method is evaluated using two different cases considering both disperse and tightly packed particles. The data show that the method conserves porosity data, is reasonably grid-independent and can generate a relatively smooth porosity field. The new method can more accurately predict the fluid-particle interactive force for polydisperse particle system than alternative methods that have been implemented in unresolved CFD-DEM codes.