A machine-learning supported multi-scale LBM-TPM model of unsaturated, anisotropic, and deformable porous materials

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Autoren

  • Mohamad Chaaban
  • Yousef Heider
  • Wai Ching Sun
  • Bernd Markert

Externe Organisationen

  • Rheinisch-Westfälische Technische Hochschule Aachen (RWTH)
  • Columbia University
Forschungs-netzwerk anzeigen

Details

OriginalspracheEnglisch
Seiten (von - bis)889-910
Seitenumfang22
FachzeitschriftInternational Journal for Numerical and Analytical Methods in Geomechanics
Jahrgang48
Ausgabenummer4
PublikationsstatusVeröffentlicht - 13 Feb. 2024

Abstract

The purpose of this paper is to investigate the utilization of artificial neural networks (ANNs) in learning models that address the nonlinear anisotropic flow and hysteresis retention behavior of deformable porous materials. Herein, the micro-geometries of various networks of porous Bentheimer Sandstones subjected to several degrees of strain from the literature are considered. For the generation of the database required for the training, validation, and testing of the machine learning (ML) models, single-phase and biphasic lattice Boltzmann (LB) simulations are performed. The anisotropic nature of the intrinsic permeability is investigated for the single-phase LB simulations. Thereafter, the database contains the computed average fluid velocities versus the pressure gradients. In this database, the range of applied fluid pressure gradients includes Darcy as well as non-Darcy flows. The generated output from the single-phase flow simulations is implemented in a feed-forward neural network, representing a path-independent informed graph-based model. Concerning the two-phase LB simulations, the Shan-Chen multiphase LB model is used to generate the retention curves of the cyclic drying/wetting processes in the deformed porous networks. Consequently, two different ML path-dependent approaches, that is, 1D convolutional neural network and the recurrent neural network, are used to model the biphasic flow through the deformable porous materials. A comparison in terms of accuracy and speed of training between the two approaches is presented. Conclusively, the outcomes of the papers show the capability of the ML models in representing constitutive relations for permeability and hysteretic retention curves accurately and efficiently.

ASJC Scopus Sachgebiete

Zitieren

A machine-learning supported multi-scale LBM-TPM model of unsaturated, anisotropic, and deformable porous materials. / Chaaban, Mohamad; Heider, Yousef; Sun, Wai Ching et al.
in: International Journal for Numerical and Analytical Methods in Geomechanics, Jahrgang 48, Nr. 4, 13.02.2024, S. 889-910.

Publikation: Beitrag in FachzeitschriftArtikelForschungPeer-Review

Download
@article{c065ca6262334a42b33feb4141476163,
title = "A machine-learning supported multi-scale LBM-TPM model of unsaturated, anisotropic, and deformable porous materials",
abstract = "The purpose of this paper is to investigate the utilization of artificial neural networks (ANNs) in learning models that address the nonlinear anisotropic flow and hysteresis retention behavior of deformable porous materials. Herein, the micro-geometries of various networks of porous Bentheimer Sandstones subjected to several degrees of strain from the literature are considered. For the generation of the database required for the training, validation, and testing of the machine learning (ML) models, single-phase and biphasic lattice Boltzmann (LB) simulations are performed. The anisotropic nature of the intrinsic permeability is investigated for the single-phase LB simulations. Thereafter, the database contains the computed average fluid velocities versus the pressure gradients. In this database, the range of applied fluid pressure gradients includes Darcy as well as non-Darcy flows. The generated output from the single-phase flow simulations is implemented in a feed-forward neural network, representing a path-independent informed graph-based model. Concerning the two-phase LB simulations, the Shan-Chen multiphase LB model is used to generate the retention curves of the cyclic drying/wetting processes in the deformed porous networks. Consequently, two different ML path-dependent approaches, that is, 1D convolutional neural network and the recurrent neural network, are used to model the biphasic flow through the deformable porous materials. A comparison in terms of accuracy and speed of training between the two approaches is presented. Conclusively, the outcomes of the papers show the capability of the ML models in representing constitutive relations for permeability and hysteretic retention curves accurately and efficiently.",
keywords = "anisotropic permeability, convolutional neural network, hysteretic retention curve, lattice Boltzmann method, multiphase fluid flow, recurrent neural network",
author = "Mohamad Chaaban and Yousef Heider and Sun, {Wai Ching} and Bernd Markert",
note = "Funding Information: The second author, Y. Heider, would like to gratefully thank the German Research Foundation (DFG) for the support of the project “Multi‐field continuum modeling of two‐fluid‐filled porous media fracture augmented by microscale‐based machine‐learning material laws”, grant number 458375627. ",
year = "2024",
month = feb,
day = "13",
doi = "10.1002/nag.3668",
language = "English",
volume = "48",
pages = "889--910",
journal = "International Journal for Numerical and Analytical Methods in Geomechanics",
issn = "0363-9061",
publisher = "John Wiley and Sons Ltd",
number = "4",

}

Download

TY - JOUR

T1 - A machine-learning supported multi-scale LBM-TPM model of unsaturated, anisotropic, and deformable porous materials

AU - Chaaban, Mohamad

AU - Heider, Yousef

AU - Sun, Wai Ching

AU - Markert, Bernd

N1 - Funding Information: The second author, Y. Heider, would like to gratefully thank the German Research Foundation (DFG) for the support of the project “Multi‐field continuum modeling of two‐fluid‐filled porous media fracture augmented by microscale‐based machine‐learning material laws”, grant number 458375627.

PY - 2024/2/13

Y1 - 2024/2/13

N2 - The purpose of this paper is to investigate the utilization of artificial neural networks (ANNs) in learning models that address the nonlinear anisotropic flow and hysteresis retention behavior of deformable porous materials. Herein, the micro-geometries of various networks of porous Bentheimer Sandstones subjected to several degrees of strain from the literature are considered. For the generation of the database required for the training, validation, and testing of the machine learning (ML) models, single-phase and biphasic lattice Boltzmann (LB) simulations are performed. The anisotropic nature of the intrinsic permeability is investigated for the single-phase LB simulations. Thereafter, the database contains the computed average fluid velocities versus the pressure gradients. In this database, the range of applied fluid pressure gradients includes Darcy as well as non-Darcy flows. The generated output from the single-phase flow simulations is implemented in a feed-forward neural network, representing a path-independent informed graph-based model. Concerning the two-phase LB simulations, the Shan-Chen multiphase LB model is used to generate the retention curves of the cyclic drying/wetting processes in the deformed porous networks. Consequently, two different ML path-dependent approaches, that is, 1D convolutional neural network and the recurrent neural network, are used to model the biphasic flow through the deformable porous materials. A comparison in terms of accuracy and speed of training between the two approaches is presented. Conclusively, the outcomes of the papers show the capability of the ML models in representing constitutive relations for permeability and hysteretic retention curves accurately and efficiently.

AB - The purpose of this paper is to investigate the utilization of artificial neural networks (ANNs) in learning models that address the nonlinear anisotropic flow and hysteresis retention behavior of deformable porous materials. Herein, the micro-geometries of various networks of porous Bentheimer Sandstones subjected to several degrees of strain from the literature are considered. For the generation of the database required for the training, validation, and testing of the machine learning (ML) models, single-phase and biphasic lattice Boltzmann (LB) simulations are performed. The anisotropic nature of the intrinsic permeability is investigated for the single-phase LB simulations. Thereafter, the database contains the computed average fluid velocities versus the pressure gradients. In this database, the range of applied fluid pressure gradients includes Darcy as well as non-Darcy flows. The generated output from the single-phase flow simulations is implemented in a feed-forward neural network, representing a path-independent informed graph-based model. Concerning the two-phase LB simulations, the Shan-Chen multiphase LB model is used to generate the retention curves of the cyclic drying/wetting processes in the deformed porous networks. Consequently, two different ML path-dependent approaches, that is, 1D convolutional neural network and the recurrent neural network, are used to model the biphasic flow through the deformable porous materials. A comparison in terms of accuracy and speed of training between the two approaches is presented. Conclusively, the outcomes of the papers show the capability of the ML models in representing constitutive relations for permeability and hysteretic retention curves accurately and efficiently.

KW - anisotropic permeability

KW - convolutional neural network

KW - hysteretic retention curve

KW - lattice Boltzmann method

KW - multiphase fluid flow

KW - recurrent neural network

UR - http://www.scopus.com/inward/record.url?scp=85178933061&partnerID=8YFLogxK

U2 - 10.1002/nag.3668

DO - 10.1002/nag.3668

M3 - Article

AN - SCOPUS:85178933061

VL - 48

SP - 889

EP - 910

JO - International Journal for Numerical and Analytical Methods in Geomechanics

JF - International Journal for Numerical and Analytical Methods in Geomechanics

SN - 0363-9061

IS - 4

ER -