Predicting Effective Diffusivity of Porous Media from Images by Deep Learning

  title={Predicting Effective Diffusivity of Porous Media from Images by Deep Learning},
  author={Haiyi Wu and Wen-Zhen Fang and Qinjun Kang and Wenquan Tao and Rui Qiao},
  journal={Scientific Reports},
We report the application of machine learning methods for predicting the effective diffusivity (De) 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… 

Convolutional neural network based prediction of effective diffusivity from microscope images

This study presents a method to predict the effective diffusivity of porous media from a limited set of scanning electron microscope images using deep learning. The electrodeposited nickel oxide film

Estimating Dispersion Coefficient in Flow Through Heterogeneous Porous Media by a Deep Convolutional Neural Network

A novel deep convolutional neural network (DCNN) for estimating longitudinal dispersion coefficient DL in flow through heterogeneous porous media and its predictions are in excellent agreement with the data not used in the training.

Point-Cloud Deep Learning of Porous Media for Permeability Prediction

A novel deep learning framework for predicting permeability of porous media from their digital images that model the boundary between solid matrix and pore spaces as point clouds and feed them as inputs to a neural network based on the PointNet architecture.

Estimating relative diffusion from 3D micro-CT images using CNNs

In the past several years, convolutional neural networks (CNNs) have proven their capability to predict characteristic quantities in porous media research directly from pore-space geometries. Due to

Prediction and Optimization of Ion Transport Characteristics in Nanoparticle-Based Electrolytes Using Convolutional Neural Networks.

This study develops a convolutional neural network model to predict the diffusivity of cations in nanoparticle-based electrolytes and uses it to identify the characteristics of morphologies that exhibit optimal transport properties, and investigates the ability of the CNN model to quantitatively account for the influence of interparticle spatial correlations on cation diffusivities.

Simulating fluid flow in complex porous materials by integrating the governing equations with deep-layered machines

An ML approach that incorporates mass conservation and the Navier–Stokes equations in its learning process is introduced and it is shown that the approach provides highly accurate predictions for the flow properties of porous media at all other times and spatial locations, while reducing the computation time.

Large-Scale Statistical Learning for Mass Transport Prediction in Porous Materials Using 90,000 Artificially Generated Microstructures

A vast database comprising 90,000 microstructures drawn from nine different stochastic models, and compute their effective diffusivity and permeability as well as various microstructural descriptors, establishing microstructure-property relationships using analytical prediction formulas, artificial (fully-connected) neural networks, and convolutional neural networks.



Mesoscopic predictions of the effective thermal conductivity for microscale random porous media.

By using the present lattice Boltzmann algorithm along with the structure generating tool QSGS, the effective thermal conductivities of porous media with multiphase structure and stochastic complex geometries are predicted, without resorting to any empirical parameters determined case by case.

Nanoscale simulation of shale transport properties using the lattice Boltzmann method: permeability and diffusivity

It is found that most of the values of correction factor fall in the slip and transition regime, with no Darcy flow regime observed, indicating Knudsen diffusion always plays a role on shale gas transport mechanisms in the reconstructed shales.

Generalized lattice Boltzmann model for flow through tight porous media with Klinkenberg's effect.

Simulations performed in a heterogeneous porous medium with components of different porosity and permeability indicate that Klinkenberg's effect plays a significant role on fluid flow in low-permeability porous media, and it is more pronounced as the Knudsen number increases.

Adam: A Method for Stochastic Optimization

This work introduces Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments, and provides a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework.

Automatic Detection of Cerebral Microbleeds From MR Images via 3D Convolutional Neural Networks

This paper proposes a novel automatic method to detect CMBs from magnetic resonance (MR) images by exploiting the 3D convolutional neural network (CNN), outperforming previous methods using low-level descriptors or 2D CNNs by a significant margin.

A dual-tube model for gas dynamics in fractured nanoporous shale formations

Abstract Gas flow through fractured nanoporous shale formations is complicated by a hierarchy of structural features (ranging from nanopores to microseismic and hydraulic fractures) and by several