A Voronoi tessellation-based CFD-DEM porosity model for fluidized bed modelling

Abstract

The porosity model is a critical component of CFD-DEM calculations, particularly in scenarios where the CFD mesh size cannot be adjusted flexibly. Voronoi tessellation has been demonstrated as an efficient method for estimating the porosity of individual particles. This study introduces a flexible implementation of the Voronoi porosity model, extending its application to broader chemical engineering scenarios and enabling large-scale computations for irregular devices. The model leverages CFD mesh cell nodes and the functions in Voro++ package, addressing previous limitations with point cloud generation. Evaluations across various particle size distributions and fluidised beds show that the model generates smooth porosity fields, remains independent of mesh grid density, and significantly improves drag force calculation accuracy in polydisperse particle systems. Although computational costs may rise due to frequent Voronoi tessellation updates, reducing the update frequency can help reduce these costs.