Statistical effective diffusivity estimation in porous media using an integrated on-site imaging workflow for synchrotron users
Statistical effective diffusivity estimation in porous media using an integrated on-site imaging workflow for synchrotron users
Transport in porous media plays an essential role for many physical, engineering, biological and environmental processes. Novel synchrotron imaging techniques and image-based models have enabled more robust quantification of geometric structures that influence transport through the pore space. However, image-based modelling is computationally expensive, and end users often require, while conducting imaging campaign, fast and agile bulk-scale effective parameter estimates that account for the pore-scale details. In this manuscript we enhance a pre-existing image-based model solver known as OpenImpala to estimate bulk-scale effective transport parameters. In particular, the boundary conditions and equations in OpenImpala were modified in order to estimate the effective diffusivity in an imaged system/geometry via a formal multi-scale homogenisation expansion. Estimates of effective pore space diffusivity were generated for a range of elementary volume sizes to estimate when the effective diffusivity values begin to converge to a single value. Results from OpenImpala were validated against a commercial finite element method package COMSOL Multiphysics (abbreviated as COMSOL). Results showed that the effective diffusivity values determined with OpenImpala were similar to those estimated by COMSOL. Tests on larger domains comparing a full image-based model to a homogenised (geometrically uniform) domain that used the effective diffusivity parameters showed differences below 2 % error, thus verifying the accuracy of the effective diffusivity estimates. Finally, we compared OpenImpala’s parallel computing speeds to COMSOL. OpenImpala consistently ran simulations within fractions of minutes, which was two orders of magnitude faster than COMSOL providing identical supercomputing specifications. In conclusion, we demonstrated OpenImpala’s utility as part of an on-site tomography processing pipeline allowing for fast and agile assessment of porous media processes and to guide imaging campaigns while they are happening at synchrotron beamlines.
Homogenisation, Image-based modelling, Numerical methods, Porous media
71-88
Le Houx, James
42aaf87b-a78e-4d1b-ae8a-36cbee4ea614
Ruiz, Siul Aljadi
d79b3b82-7c0d-47cc-9616-11d29e6a41bd
McKay Fletcher, Daniel
83a4bf44-f5c4-442b-8403-479c05fd98ad
Ahmeds, Sharif
38d9f9d8-2306-4705-9ba9-fb65b2296c3d
Roose, Tiina
3581ab5b-71e1-4897-8d88-59f13f3bccfe
26 July 2023
Le Houx, James
42aaf87b-a78e-4d1b-ae8a-36cbee4ea614
Ruiz, Siul Aljadi
d79b3b82-7c0d-47cc-9616-11d29e6a41bd
McKay Fletcher, Daniel
83a4bf44-f5c4-442b-8403-479c05fd98ad
Ahmeds, Sharif
38d9f9d8-2306-4705-9ba9-fb65b2296c3d
Roose, Tiina
3581ab5b-71e1-4897-8d88-59f13f3bccfe
Le Houx, James, Ruiz, Siul Aljadi, McKay Fletcher, Daniel, Ahmeds, Sharif and Roose, Tiina
(2023)
Statistical effective diffusivity estimation in porous media using an integrated on-site imaging workflow for synchrotron users.
Transport in Porous Media, 150 (1), .
(doi:10.1007/s11242-023-01993-7).
Abstract
Transport in porous media plays an essential role for many physical, engineering, biological and environmental processes. Novel synchrotron imaging techniques and image-based models have enabled more robust quantification of geometric structures that influence transport through the pore space. However, image-based modelling is computationally expensive, and end users often require, while conducting imaging campaign, fast and agile bulk-scale effective parameter estimates that account for the pore-scale details. In this manuscript we enhance a pre-existing image-based model solver known as OpenImpala to estimate bulk-scale effective transport parameters. In particular, the boundary conditions and equations in OpenImpala were modified in order to estimate the effective diffusivity in an imaged system/geometry via a formal multi-scale homogenisation expansion. Estimates of effective pore space diffusivity were generated for a range of elementary volume sizes to estimate when the effective diffusivity values begin to converge to a single value. Results from OpenImpala were validated against a commercial finite element method package COMSOL Multiphysics (abbreviated as COMSOL). Results showed that the effective diffusivity values determined with OpenImpala were similar to those estimated by COMSOL. Tests on larger domains comparing a full image-based model to a homogenised (geometrically uniform) domain that used the effective diffusivity parameters showed differences below 2 % error, thus verifying the accuracy of the effective diffusivity estimates. Finally, we compared OpenImpala’s parallel computing speeds to COMSOL. OpenImpala consistently ran simulations within fractions of minutes, which was two orders of magnitude faster than COMSOL providing identical supercomputing specifications. In conclusion, we demonstrated OpenImpala’s utility as part of an on-site tomography processing pipeline allowing for fast and agile assessment of porous media processes and to guide imaging campaigns while they are happening at synchrotron beamlines.
Text
s11242-023-01993-7
- Version of Record
More information
Accepted/In Press date: 3 July 2023
e-pub ahead of print date: 26 July 2023
Published date: 26 July 2023
Additional Information:
Funding Information:
This work was carried out with the support of Diamond Light Source, at the Dual Imaging And Diffraction (DIAD) beamline as part of the Ada Lovelace Centre (ALC) STFC project, CANVAS-NXtomo, ContAiNerised Voxel-bAsed Simulation of Neutron and X-ray Tomography data. This work was part funded by the EPSRC prosperity partnership with Imperial College, INFUSE, Interface with the Future - Underpinning Science to Support the Energy transition EP/V038044/1. This work was also part funded by the ERC Consolidator grant 646809 (Data Intensive Modelling of the Rhizosphere Processes), BBSRC SARIC BB/P004180/1, BBSRC SARISA BB/L025620/1 and EPSRC EP/M020355/1.
Funding Information:
This work was carried out with the support of Diamond Light Source, at the Dual Imaging And Diffraction (DIAD) beamline as part of the Ada Lovelace Centre (ALC) STFC project, CANVAS-NXtomo, ContAiNerised Voxel-bAsed Simulation of Neutron and X-ray Tomography data. This work was part funded by the EPSRC prosperity partnership with Imperial College, INFUSE, Interface with the Future - Underpinning Science to Support the Energy transition EP/V038044/1. This work was also part funded by the ERC Consolidator grant 646809 (Data Intensive Modelling of the Rhizosphere Processes), BBSRC SARIC BB/P004180/1, BBSRC SARISA BB/L025620/1 and EPSRC EP/M020355/1.
Funding Information:
JLH was supported by EPSRC grant EP/V038044/1, and the STFC ALC project, CANVAS-NXtomo. SR, DMF and TR are funded by ERC Consolidator grant 646809 (Data Intensive Modelling of the Rhizosphere Processes). TR is also funded by, BBSRC SARIC BB/P004180/1, BBSRC SARISA BB/L025620/1 and EPSRC EP/M020355/1. DMF was also supported by the Rural and Environment Science and Analytical Services Division (SRUC-C5-1). SR was also supported by BBSRC Discovery Fellowship BB/X010147/1 (Quantifying soil biomechanics using X-Ray diffraction-imaging and physical modelling).
Publisher Copyright:
© 2023, The Author(s).
Keywords:
Homogenisation, Image-based modelling, Numerical methods, Porous media
Identifiers
Local EPrints ID: 481345
URI: http://eprints.soton.ac.uk/id/eprint/481345
ISSN: 0169-3913
PURE UUID: 1720dfce-45d1-4f00-bffe-7f477b19218f
Catalogue record
Date deposited: 23 Aug 2023 17:02
Last modified: 06 Jun 2024 01:46
Export record
Altmetrics
Contributors
Author:
James Le Houx
Author:
Daniel McKay Fletcher
Author:
Sharif Ahmeds
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
