Skip to main content
OSTI.GOVU.S. Department of Energy Office of Scientific and Technical Information
U.S. Department of Energy
Office of Scientific and Technical Information
 

Hierarchical reconstruction of 3D well-connected porous media from 2D exemplars using statistics-informed neural network

Journal Article · · Computer Methods in Applied Mechanics and Engineering
 [1];  [2];  [3];  [1];  [1];  [4];  [1]
  1. Swansea University (United Kingdom)
  2. Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
  3. Tongji University, Shanghai (China)
  4. China University of Petroleum, Qingdao (China)
+ Show Author Affiliations
The relationships between porous microstructures and transport properties are of fundamental importance in various scientific and engineering applications. Due to the intricacy, stochasticity and heterogeneity of porous media, reliable characterization and modeling of transport properties often require a complete dataset of internal microstructure samples. However, it is often an unbearable cost to acquire sufficient 3D digital microstructures by purely using microscopic imaging systems. Herein this paper presents a machine learning-based technique to hierarchically reconstruct 3D well-connected porous microstructures from one isotropic or several anisotropic low-cost 2D exemplar(s). To compactly characterize the large-scale microstructural features, a Gaussian image pyramid is built for each 2D exemplar. Local morphology patterns are collected from the Gaussian image pyramids, and then they serve as the training data to embed the 2D morphological statistics into feed-forward neural networks at multiple length levels. By using a specially-developed morphology integration scheme, the 3D morphological statistics at different levels can be inferred from the statistics-informed neural networks. Gibbs sampling is adopted to hierarchically reconstruct 3D microstructures by using multi-level 3D morphological statistics, where the large-scale, regional and local morphological patterns are statistically generated and successively added to the same 3D random field. The proposed method is tested on a series of porous media with distinct morphologies, and the statistical equivalence between the reconstructed and the real microstructures is systematically evaluated by comparing morphological descriptors and transport properties. The results demonstrate that the proposed 2D-to-3D microstructure reconstruction method is a universal and efficient approach to generating morphologically and physically realistic samples of porous media.
Research Organization:
Los Alamos National Laboratory (LANL), Los Alamos, NM (United States)
Sponsoring Organization:
USDOE National Nuclear Security Administration (NNSA); Engineering and Physical Sciences Research Council (EPSRC)
Grant/Contract Number:
89233218CNA000001
OSTI ID:
1974987
Report Number(s):
LA-UR-23-24399
Journal Information:
Computer Methods in Applied Mechanics and Engineering, Journal Name: Computer Methods in Applied Mechanics and Engineering Vol. 410; ISSN 0045-7825
Publisher:
ElsevierCopyright Statement
Country of Publication:
United States
Language:
English

References (53)

Revisiting the empirical particle‐fluid coupling model used in DEM‐CFD by high‐resolution DEM‐LBM‐IMB simulations: A 2D perspective
  • Zeng, Zhixiong; Fu, Jinlong; Feng, Y. T.
  • International Journal for Numerical and Analytical Methods in Geomechanics, Vol. 47, Issue 5 https://doi.org/10.1002/nag.3496
journal January 2023
A multiscale computational fluid dynamics approach to simulate the micro-fluidic environment within a tissue engineering scaffold with highly irregular pore geometry journal June 2019
Porous Structure Reconstruction Using Convolutional Neural Networks journal May 2018
3D Stochastic Modelling of Heterogeneous Porous Media – Applications to Reservoir Rocks journal December 2006
Permeability Description by Characteristic Length, Tortuosity, Constriction and Porosity journal April 2014
Resolution Effect: An Error Correction Model for Intrinsic Permeability of Porous Media Estimated from Lattice Boltzmann Method journal March 2020
A new three-dimensional modeling technique for studying porous media journal March 1984
Introduction to multi-layer feed-forward neural networks journal November 1997
The microstructure and electrical conductivity of aluminum alloy foams journal February 2003
A scaled conjugate gradient algorithm for fast supervised learning journal January 1993
A reconstruction technique for three-dimensional porous media using image analysis and Fourier transforms journal November 1998
Stochastic microstructure characterization and reconstruction via supervised learning journal January 2016
A multiple-point statistics algorithm for 3D pore space reconstruction from 2D images journal October 2011
Pore-scale imaging and modelling journal January 2013
A lattice Boltzmann study on the impact of the geometrical properties of porous media on the steady state relative permeabilities on two-phase immiscible flows journal September 2016
Reconstruction of 3D Microstructures from 2D Images via Transfer Learning journal November 2020
Characterization and simulation of microstructure and properties of EPS lightweight concrete journal December 2007
Porous ceramic materials by pore-forming agent method: An intermingled fractal units analysis and procedure to predict thermal conductivity journal June 2015
Compressive strength and microstructure of soft clay soil stabilized with recycled bassanite journal February 2015
GPU implementation of lattice Boltzmann method for flows with curved boundaries journal June 2012
3D numerical reconstruction of well-connected porous structure of rock using fractal algorithms journal September 2014
New algorithms for virtual reconstruction of heterogeneous microstructures journal August 2018
Analyzing effects of surface roughness, voids, and particle–matrix interfacial bonding on the failure response of a heterogeneous adhesive journal April 2019
Predicting the effective mechanical property of heterogeneous materials by image based modeling and deep learning journal April 2019
Efficient 3D reconstruction of random heterogeneous media via random process theory and stochastic reconstruction procedure journal September 2019
An end-to-end three-dimensional reconstruction framework of porous media from a single two-dimensional image based on deep learning journal August 2020
Statistical characterization and reconstruction of heterogeneous microstructures using deep neural network journal January 2021
The correlation between statistical descriptors of heterogeneous materials journal October 2021
Stochastic reconstruction of 3D microstructures from 2D cross-sectional images using machine learning-based characterization journal February 2022
Microstructure-guided deep material network for rapid nonlinear material modeling and uncertainty quantification journal August 2022
Physics-data combined machine learning for parametric reduced-order modelling of nonlinear dynamical systems in small-data regimes journal February 2023
Classification and reconstruction of three-dimensional microstructures using support vector machines journal February 2005
Descriptor-based methodology for statistical characterization and 3D reconstruction of microstructural materials journal April 2014
Random heterogeneous materials via texture synthesis journal March 2015
Improving direct physical properties prediction of heterogeneous materials from imaging data via convolutional neural network and a morphology-aware generative model journal July 2018
Slice-to-voxel stochastic reconstructions on porous media with hybrid deep generative model journal January 2021
Tortuosity of porous media: Image analysis and physical simulation journal January 2021
Pore-scale study of reactive transport processes in catalyst layer agglomerates of proton exchange membrane fuel cells journal May 2019
Preparation and properties of porous alumina ceramics with oriented cylindrical pores produced by an extrusion method journal January 2006
Effects of pore shape and porosity on the properties of porous PZT 95/5 ceramics journal January 2007
Hybrid approach combining multiple characterization techniques and simulations for microstructural analysis of proton exchange membrane fuel cell electrodes journal March 2017
3D-PMRNN: Reconstructing three-dimensional porous media from the two-dimensional image with recurrent neural network journal January 2022
TauFactor: An open-source application for calculating tortuosity factors from tomographic data journal January 2016
Review on Scale Dependent Characterization of the Microstructure of Porous Media journal January 2002
Process Based Reconstruction of Sandstones and Prediction of Transport Properties journal January 2002
Permeability prediction from thin sections: 3D reconstruction and Lattice-Boltzmann flow simulation journal January 2004
Generating three-dimensional structures from a two-dimensional slice with generative adversarial network-based dimensionality expansion journal April 2021
Reconstructing random media. II. Three-dimensional media from two-dimensional cuts journal July 1998
Prediction of permeability for porous media reconstructed using multiple-point statistics journal December 2004
Reconstruction of three-dimensional porous media using a single thin section journal June 2012
Characterization and reconstruction of 3D stochastic microstructures via supervised learning: CHARACTERIZATION AND RECONSTRUCTION OF 3D STOCHASTIC MICROSTRUCTURES journal July 2016
Advances in multiscale numerical and experimental approaches for multiphysics problems in porous media journal June 2021
Effects of pore connectivity and water saturation on matrix permeability of deep gas shale journal January 2022

Figures / Tables (37)

Figure 1(p. 3)figure Figure 1
Figure 2(p. 4)figure Figure 2
Figure 3(p. 5)figure Figure 3
Figure 4(p. 6)figure Figure 4
Figure 5(p. 7)figure Figure 5
Figure 6(p. 8)figure Figure 6
Figure 7(p. 10)figure Figure 7
Figure 8(p. 12)figure Figure 8
Figure 9(p. 18)figure Figure 9
Table 1(p. 21)table Table 1
Figure 10(p. 22)figure Figure 10
Figure 11(p. 23)figure Figure 11
Figure 12(p. 23)figure Figure 12
Figure 13(p. 24)figure Figure 13
Figure 14(p. 25)figure Figure 14
Figure 15(p. 26)figure Figure 15
Figure 16(p. 26)figure Figure 16
Figure 17(p. 27)figure Figure 17
Figure 18(p. 27)figure Figure 18
Figure 19(p. 28)figure Figure 19
Figure 20(p. 28)figure Figure 20
Figure 21(p. 29)figure Figure 21
Figure 22(p. 30)figure Figure 22
Figure 23(p. 30)figure Figure 23
Figure 24(p. 31)figure Figure 24
Figure 25(p. 32)figure Figure 25
Figure 26(p. 33)figure Figure 26
Figure 27(p. 33)figure Figure 27
Figure 28(p. 34)figure Figure 28
Figure 29(p. 34)figure Figure 29
Figure 30(p. 35)figure Figure 30
Figure 31(p. 36)figure Figure 31
Figure 32(p. 37)figure Figure 32
Figure 33(p. 37)figure Figure 33
Figure 34(p. 38)figure Figure 34
Figure 35(p. 39)figure Figure 35
Figure 36(p. 39)figure Figure 36

Similar Records

An Algorithm to Generate Synthetic 3D Microstructures from 2D Exemplars
Journal Article · Tue Oct 15 00:00:00 EDT 2019 · JOM. Journal of the Minerals, Metals & Materials Society · OSTI ID:1617323
Microscopic imaging of porous media with X-ray computer tomography
Journal Article · Wed Sep 01 00:00:00 EDT 1993 · SPE Formation Evaluation (Society of Petroleum Engineers); (United States) · OSTI ID:5940482

Related Subjects

42 ENGINEERING
hierarchical reconstruction
long-distance correlation
porous microstructure
small data
statistics-informed neural network
transport property