Complex network analysis of pore structures in monodisperse granular materials with varied grain shapes

TL;DR

This paper presents a novel framework combining DEM simulations, image processing, and complex network theory to analyze how particle shape affects pore structures and hydraulic behavior in granular materials.

Contribution

It introduces an integrated approach to link particle morphology with pore network characteristics using advanced modeling and network analysis techniques.

Findings

Particle shape influences pore and throat size distributions.

Pore network metrics correlate with hydraulic transport properties.

Morphological variations affect pore network anisotropy.

Abstract

Understanding how pore structure influences flow and transport behaviour in granular materials is essential for addressing a wide range of geotechnical, hydraulic, and environmental challenges. These processes are largely shaped by the microscopic arrangement of particles and interconnections between pores within the material. However, detailed insights considering granular assemblies with diversified grain shapes remain scarce. This study introduces a comprehensive framework incorporating Discrete Element Method (DEM) simulations, image processing, pore-network modelling, and complex network theory to investigate the links between particle morphology and their hydraulic behaviours. Mono-disperse assemblies of natural sand particles with varied shapes are constructed in DEM, and pore networks are extracted through image processing and pore-network modelling. Complex network analysis is then applied to calculate structural metrics that reveal intrinsic relationships between pore microstructures and hydraulic properties. Our results demonstrate that particle morphology significantly impacts pore network characteristics, including pore and throat sizes, closeness centrality, pore structure anisotropy, providing valuable insights into how pore structure influence transport properties.