Publications

Detailed Information

GPU-based SPH-DEM Method to Examine the Three-Phase Hydrodynamic Interactions between Multiphase Flow and Solid Particles

Cited 8 time in Web of Science Cited 9 time in Scopus
Authors

Jo, Young Beom; Park, So-Hyun; Yoo, Hee Sang; Kim, Eung Soo

Issue Date
2022-08
Publisher
Pergamon Press Ltd.
Citation
International Journal of Multiphase Flow, Vol.153, p. 104125
Abstract
Particulate flows in a gas-liquid multiphase mixture (i.e., three-phase flows) are frequently encountered in various fields as well as the chemical engineering, resource engineering, ocean engineering, and nuclear engineering domains. Thus, the numerical modeling for such three-phase flows is essential to evaluate and enhance the fluidic design of the existing industrial devices or increase the engineering safety/efficiency. In terms of the solid particle phase, the discrete element method (DEM) has demonstrated outstanding capability in numerically modeling solid-solid interactions. For multiphase CFD model, several numerical techniques, ranging from the Eulerian-Eulerian two-fluid model to a fully resolved DNS solver, have been coupled with the DEM to model the three-phase flow. Most of the existing research on three-phase particulate flows was based on coupled grid-based CFD and DEM techniques. Considering that many three-phase particulate flows in industries are driven by a vapor/gas phase that forms a sharp interface with the liquid, coupling DEM with particle-based Lagrangian CFD methods, which can handle multi-phase flows without any interface capturing or associated correlations, can also be a good option or alternative for modeling such three-phase flows. In this respect, a three-phase flow solver was established in this study by coupling the DEM with a Lagrangian smoothed particle hydrodynamics (SPH) model while adopting the unresolved phase coupling between the two techniques. In order to overcome the low computational efficiency of these two numerical methods, GPU-based acceleration is adopted for speedup of SPH-DEM coupled solver due to its high compatibility for Lagrangian methods offered by thousands of GPU processors. The results of a scalability analysis indicated that the accelerated version exhibited an enhanced performance especially when many Lagrangian nodes are used in simulation. Several benchmark simulations were conducted to demonstrate the capability and applicability of the SPH-DEM solver in various multi-phase applications associated with particulate flows. The results of numerical simulations validated the feasibility of the momentum exchange model, and the potential and applicability of the SPH-DEM coupled solver for threephase flow was demonstrated through a three-phase SPH-DEM simulation on the gas-driven leveling behavior of a particulate bed.
ISSN
0301-9322
URI
https://hdl.handle.net/10371/184347
DOI
https://doi.org/10.1016/j.ijmultiphaseflow.2022.104125
Files in This Item:
There are no files associated with this item.
Appears in Collections:

Altmetrics

Item View & Download Count

  • mendeley

Items in S-Space are protected by copyright, with all rights reserved, unless otherwise indicated.

Share