Coupled CFD-DEM Parallel Model to Describe Fines Migration Behavior at the Pore-Scale

Abstract

Transport of fine particles in porous media has attracted considerable interest over the past few decades. In various fields of science and engineering, on-site remediation strategies, enhanced transport of contaminants into aquifers, artificial recharging of groundwater aquifers, and permeability of oil and gas reservoirs have become topics of concern. Much research projects focused on the development of models that can predict their behavior on a pore-scale level. The model described here uses a Computational Fluid Dynamics - Discrete Element Method (CFD-DEM) parallel model to simulate fine particle migration and interactions in porous media. The model was parallelized using Message Passing Interface (MPI) for communication between processors to achieve faster computation. Model validation was performed by confirming the fluid flow separately using two geometries. Then, the coupled CFD-DEM model was validated by comparing its results to a micromodel experiment with the equivalent geometry. Comparison was done by utilizing different computer vision techniques to capture the behavior of the particles throughout the micro-model experiment. The CFD-DEM was able to capture the particles transport in porous media while maintaining low computational cost with the use of parallelization. The open-source library OpenCV showed great potential to be used for image and video processing for the results captured from micromodel experiments, which could be used for coupled model validations as well as data analysis in future research.