Abstract:Abstract The permeability of complex porous materials is of interest to many engineering disciplines. This quantity can be obtained via direct flow simulation, which provides the most accurate results, but is very computationally expensive. In particular, the simulation convergence time scales poorly as the simulation domains become less porous or more heterogeneous. Semi-analytical models that rely on averaged structural properties (i.e., porosity and tortuosity) have been proposed, but these features only partly summarize the domain, resulting in limited applicability. On the other hand, data-driven machine learning approaches have shown great promise for building more general models by virtue of accounting for the spatial arrangement of the domains’ solid boundaries. However, prior approaches building on the convolutional neural network (ConvNet) literature concerning 2D image recognition problems do not scale well to the large 3D domains required to obtain a representative elementary volume (REV). As such, most prior work focused on homogeneous samples, where a small REV entails that the global nature of fluid flow could be mostly neglected, and accordingly, the memory bottleneck of addressing 3D domains with ConvNets was side-stepped. Therefore, important geometries such as fractures and vuggy domains could not be modeled properly. In this work, we address this limitation with a general multiscale deep learning model that is able to learn from porous media simulation data. By using a coupled set of neural networks that view the domain on different scales, we enable the evaluation of large ( $$>512^3$$ > 512 3 ) images in approximately one second on a single graphics processing unit. This model architecture opens up the possibility of modeling domain sizes that would not be feasible using traditional direct simulation tools on a desktop computer. We validate our method with a laminar fluid flow case using vuggy samples and fractures. As a result of viewing the entire domain at once, our model is able to perform accurate prediction on domains exhibiting a large degree of heterogeneity. We expect the methodology to be applicable to many other transport problems where complex geometries play a central role.

Simulations of Disordered Matter in 3D with the Morphological Autoregressive Protocol (MAP) and Convolutional Neural Networks

Simulations of disordered matter in 3D with the morphological autoregressive protocol (MAP) and convolutional neural networks

Machine learning for predictive condensed-phase simulation

Random-Walk Metaball-Imaging Discrete Element Lattice Boltzmann Method for 3D Solute Transport in Fluid-Particle Systems with Complex Granular Morphologies.

Perspective: Atomistic Simulations of Water and Aqueous Systems with Machine Learning Potentials

Computationally Efficient Multiscale Neural Networks Applied to Fluid Flow in Complex 3D Porous Media

DiffMD: A Geometric Diffusion Model for Molecular Dynamics Simulations

Sliced Wasserstein Distance-Guided Three-Dimensional Porous Media Reconstruction Based on Cycle-Consistent Adversarial Network and Few-Shot Learning

Prediction of Water Transport Properties on an Anisotropic Wetting Surface Via Deep Learning.

Modeling Liquid Water by Climbing Up Jacob's Ladder in Density Functional Theory Facilitated by Using Deep Neural Network Potentials

Transferable Water Potentials Using Equivariant Neural Networks

Enhanced Deep Potential Model for Fast and Accurate Molecular Dynamics; Application to the Hydrated Electron

A Neural Material Point Method for Particle-based Simulations

Thermodynamics of Water and Ice from a Fast and Scalable First-Principles Neuroevolution Potential

Molecular Autonomous Pathfinder Using Deep Reinforcement Learning

Machine learning potentials for complex aqueous systems made simple

A kinetic Monte Carlo approach to study fluid transport in pore networks.

Extending the atomic decomposition and many-body representation, a chemistry-motivated monomer-centered approach for machine learning potentials

Coarse Graining of Multiple Water Molecules with Intelligent Dissipative Particle Dynamics

Breaking the size limitation of nonadiabatic molecular dynamics in condensed matter systems with local descriptor machine learning

Graph Convolutional Neural Networks for (QM)ML/MM Molecular Dynamics Simulations