skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Predicting Effective Diffusivity of Porous Media from Images by Deep Learning

Abstract

Abstract We report the application of machine learning methods for predicting the effective diffusivity ( D e ) of two-dimensional porous media from images of their structures. Pore structures are built using reconstruction methods and represented as images, and their effective diffusivity is computed by lattice Boltzmann (LBM) simulations. The datasets thus generated are used to train convolutional neural network (CNN) models and evaluate their performance. The trained model predicts the effective diffusivity of porous structures with computational cost orders of magnitude lower than LBM simulations. The optimized model performs well on porous media with realistic topology, large variation of porosity (0.28–0.98), and effective diffusivity spanning more than one order of magnitude (0.1 ≲ D e < 1), e.g., >95% of predicted D e have truncated relative error of <10% when the true D e is larger than 0.2. The CNN model provides better prediction than the empirical Bruggeman equation, especially for porous structure with small diffusivity. The relative error of CNN predictions, however, is rather high for structures with D e  < 0.1. To address this issue, the porosity of porous structures is encoded directly into the neural network but the performance is enhanced marginally. Further improvement, i.e.,more » 70% of the CNN predictions for structures with true D e  < 0.1 have relative error <30%, is achieved by removing trapped regions and dead-end pathways using a simple algorithm. These results suggest that deep learning augmented by field knowledge can be a powerful technique for predicting the transport properties of porous media. Directions for future research of machine learning in porous media are discussed based on detailed analysis of the performance of CNN models in the present work.« less

Authors:
; ; ORCiD logo; ;
Publication Date:
Research Org.:
Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
Sponsoring Org.:
USDOE Laboratory Directed Research and Development (LDRD) Program
OSTI Identifier:
1619666
Alternate Identifier(s):
OSTI ID: 1599031
Report Number(s):
LA-UR-19-23183
Journal ID: ISSN 2045-2322; 20387; PII: 56309
Grant/Contract Number:  
89233218CNA000001
Resource Type:
Journal Article: Published Article
Journal Name:
Scientific Reports
Additional Journal Information:
Journal Name: Scientific Reports Journal Volume: 9 Journal Issue: 1; Journal ID: ISSN 2045-2322
Publisher:
Nature Publishing Group
Country of Publication:
United Kingdom
Language:
English
Subject:
42 ENGINEERING; Earth Sciences; Material Science

Citation Formats

Wu, Haiyi, Fang, Wen-Zhen, Kang, Qinjun, Tao, Wen-Quan, and Qiao, Rui. Predicting Effective Diffusivity of Porous Media from Images by Deep Learning. United Kingdom: N. p., 2019. Web. doi:10.1038/s41598-019-56309-x.
Wu, Haiyi, Fang, Wen-Zhen, Kang, Qinjun, Tao, Wen-Quan, & Qiao, Rui. Predicting Effective Diffusivity of Porous Media from Images by Deep Learning. United Kingdom. https://doi.org/10.1038/s41598-019-56309-x
Wu, Haiyi, Fang, Wen-Zhen, Kang, Qinjun, Tao, Wen-Quan, and Qiao, Rui. 2019. "Predicting Effective Diffusivity of Porous Media from Images by Deep Learning". United Kingdom. https://doi.org/10.1038/s41598-019-56309-x.
@article{osti_1619666,
title = {Predicting Effective Diffusivity of Porous Media from Images by Deep Learning},
author = {Wu, Haiyi and Fang, Wen-Zhen and Kang, Qinjun and Tao, Wen-Quan and Qiao, Rui},
abstractNote = {Abstract We report the application of machine learning methods for predicting the effective diffusivity ( D e ) of two-dimensional porous media from images of their structures. Pore structures are built using reconstruction methods and represented as images, and their effective diffusivity is computed by lattice Boltzmann (LBM) simulations. The datasets thus generated are used to train convolutional neural network (CNN) models and evaluate their performance. The trained model predicts the effective diffusivity of porous structures with computational cost orders of magnitude lower than LBM simulations. The optimized model performs well on porous media with realistic topology, large variation of porosity (0.28–0.98), and effective diffusivity spanning more than one order of magnitude (0.1 ≲ D e < 1), e.g., >95% of predicted D e have truncated relative error of <10% when the true D e is larger than 0.2. The CNN model provides better prediction than the empirical Bruggeman equation, especially for porous structure with small diffusivity. The relative error of CNN predictions, however, is rather high for structures with D e  < 0.1. To address this issue, the porosity of porous structures is encoded directly into the neural network but the performance is enhanced marginally. Further improvement, i.e., 70% of the CNN predictions for structures with true D e  < 0.1 have relative error <30%, is achieved by removing trapped regions and dead-end pathways using a simple algorithm. These results suggest that deep learning augmented by field knowledge can be a powerful technique for predicting the transport properties of porous media. Directions for future research of machine learning in porous media are discussed based on detailed analysis of the performance of CNN models in the present work.},
doi = {10.1038/s41598-019-56309-x},
url = {https://www.osti.gov/biblio/1619666}, journal = {Scientific Reports},
issn = {2045-2322},
number = 1,
volume = 9,
place = {United Kingdom},
year = {2019},
month = {12}
}

Journal Article:
Free Publicly Available Full Text
Publisher's Version of Record at https://doi.org/10.1038/s41598-019-56309-x

Citation Metrics:
Cited by: 3 works
Citation information provided by
Web of Science

Save / Share:

Works referenced in this record:

Seeing permeability from images: fast prediction with convolutional neural networks
journal, September 2018


Automatic Detection of Cerebral Microbleeds From MR Images via 3D Convolutional Neural Networks
journal, May 2016


Gas Diffusion in Porous Media
journal, July 1959


Mesoscopic predictions of the effective thermal conductivity for microscale random porous media
journal, March 2007


Pore Scale Simulation of Transport and Electrochemical Reactions in Reconstructed PEMFC Catalyst Layers
journal, January 2010


3D Convolutional Neural Networks for Human Action Recognition
journal, January 2013


Deep learning
journal, May 2015


Fast connected-component labeling
journal, September 2009


Gradient-based learning applied to document recognition
journal, January 1998


Deep learning approaches for mining structure-property linkages in high contrast composites from simulation datasets
journal, August 2018


Material structure-property linkages using three-dimensional convolutional neural networks
journal, March 2018


Pore Architecture and Connectivity in Gas Shale
journal, February 2015


Generalized lattice Boltzmann model for flow through tight porous media with Klinkenberg's effect
journal, March 2015


The permeability of fibrous porous media
journal, June 1986


Multi-column deep neural networks for image classification
conference, June 2012


A dual-tube model for gas dynamics in fractured nanoporous shale formations
journal, October 2014


Fast Feature Pyramids for Object Detection
journal, August 2014