No Cover Image

Journal article 545 views

Effect of image scaling and segmentation in digital rock characterisation

B. D. Jones, Y. T. Feng, Yuntian Feng Orcid Logo

Computational Particle Mechanics, Volume: 3, Issue: 2, Pages: 201 - 213

Swansea University Author: Yuntian Feng Orcid Logo

Full text not available from this repository: check for access using links below.

Abstract

Digital material characterisation from microstructural geometry is an emerging field in computer simulation. For permeability characterisation, a variety of studies exist where the lattice Boltzmann method (LBM) has been used in conjunction with computed tomography (CT) imaging to simulate fluid flo...

Full description

Published in: Computational Particle Mechanics
ISSN: 2196-4378 2196-4386
Published: 2016
Online Access: Check full text

URI: https://cronfa.swan.ac.uk/Record/cronfa25671
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract: Digital material characterisation from microstructural geometry is an emerging field in computer simulation. For permeability characterisation, a variety of studies exist where the lattice Boltzmann method (LBM) has been used in conjunction with computed tomography (CT) imaging to simulate fluid flow through microscopic rock pores. While these previous works show that the technique is applicable, the use of binary image segmentation and the bounceback boundary condition results in a loss of grain surface definition when the modelled geometry is compared to the original CT image. We apply the immersed moving boundary (IMB) condition of Noble and Torczynski as a partial bounceback boundary condition which may be used to better represent the geometric definition provided by a CT image. The IMB condition is validated against published work on idealised porous geometries in both 2D and 3D. Following this, greyscale image segmentation is applied to a CT image of Diemelstadt sandstone. By varying the mapping of CT voxel densities to lattice sites, it is shown that binary image segmentation may underestimate the true permeability of the sample. A CUDA-C-based code, LBM-C, was developed specifically for this work and leverages GPU hardware in order to carry out computations.
Keywords: Lattice Boltzmann, CT imaging, Permeability, GPU computing
College: College of Engineering
Issue: 2
Start Page: 201
End Page: 213