Machine learning framework for analysis of transport through complex networks in porous, granular media: A focus on permeability

JH Van Der Linden; GA Narsilio; A Tordesillas

Journal article

Machine learning framework for analysis of transport through complex networks in porous, granular media: A focus on permeability

JH Van Der Linden, GA Narsilio, A Tordesillas

Physical Review E | AMER PHYSICAL SOC | Published : 2016

DOI: 10.1103/PhysRevE.94.022904

Abstract

We present a data-driven framework to study the relationship between fluid flow at the macroscale and the internal pore structure, across the micro- and mesoscales, in porous, granular media. Sphere packings with varying particle size distribution and confining pressure are generated using the discrete element method. For each sample, a finite element analysis of the fluid flow is performed to compute the permeability. We construct a pore network and a particle contact network to quantify the connectivity of the pores and particles across the mesoscopic spatial scales. Machine learning techniques for feature selection are employed to identify sets of microstructural properties and multiscale..

View full abstract

University of Melbourne Researchers

Guillermo Narsilio Ferrero Author

Antoinette Tordesillas Author

Related Projects (17)

Modelling and simulation of emergent behaviour in particulate assemblies under terrestrial and microgravity conditions: a focus on force propogation

Systems and processes involving particulate matter are complex, and due to inadequate understanding and modelling capabilities, rarely reach..

MULTISCALE PHENOMENA IN THE SOLID-LIQUID TRANSITION STATE OF GRANULAR MATERIAL: ANALYSIS, MODELLING AND EXPERIMENTATION

Towards generalized continuum models of granular and soil-tire interaction

Seeing the discrete in a continuum: an integrated numerical-rheological-experimental approach towards high resolution micromechanical continuum ...

Processes involving granular materials are complex and rarely reach more than 60% of the design capacity, due to inadequate understanding of..

Concept testing for a novel process for dewatering of harvested algal material

CONCEPT TESTING OF A NOVEL PROCESS FOR ENHANCED SOLID-LIQUID SEPARATION OF WASTE REPROCESSING STREAMS

Direct geothermal energy: harnessing an emerging technology

To mitigate the impacts of climate change, the demand for renewable energy technologies with low greenhouse gas (GHG) emissions is rapidly b..

Towards designing complex networks resilient to stealthy attack and cascading failure

A complex systems approach to soil informatics and taxonomies

SUPPORT TO VISIT ERDC GSL AND ITL IN VICKSBURG, MS, USA TO DISCUSS CHARACTERISATION AND MODELING OF GRANULAR SYSTEMS FROM DATA

DIRECTING TRANSMISSION PATTERNS IN GRANULAR MATERIALS FROM THE GRAIN SCALE

Knowledge discovery from transmission patterns in granular networks

CONFINED COMMINUTION AND PARTICLE FLOW: A GENERAL MODEL FOR LARGE-SCALE CANONICAL SOLUTIONS

The project integrates recent advances in continuum mechanics to develop a novel theory of comminution for large-scale problems of grain-siz..

MICROMECHANICAL MODELLING OF FAULT GOUGE DYNMICS: TOWARDS AN IMPROVED FAULT CONSTITUTIVE RELATION

The human and economic costs of geological and other particulate media related problems in Australia are staggering. These include geologic..

Topology, Structure and Functionality: Analysis, Modelling and Experimentation of Dense Granular Deformation in 2D and 3D

A complex systems approach to granular rheology: interconnecting topology, stability, dynamics and function

The response of granular materials (e.g. soil, rocks) to applied stresses and strains will be characterised in detail. Information mined fr..

Grants

Awarded by Air Force Office of Scientific Research

Funding Acknowledgements

The authors acknowledge the support of the Melbourne Energy Institute, the Australian Research Council (Grants No. FT140100227 and No. DP120104759), the U.S. Air Force (Grant No. AFOSR 15IOA059), and the U.S. Army Research Office (Grant No. W911NF-11-1-0175). We thank Dr. B. Rubinstein and Professor S. Matthai for the helpful discussions, and Dr. T. Aste for sharing the reference sample data.