Three-dimensional Voronoi Analysis of Realistic Grain Packing: an XCT Assisted Set Voronoi Tessellation Framework

Changhui Zhang,Shiwei Zhao,Jidong Zhao,Xiaowen Zhou
DOI: https://doi.org/10.1016/j.powtec.2020.10.054
IF: 5.64
2021-01-01
Powder Technology
Abstract:Voids play an important role in both transport properties and mechanical deformation of a granular material. It remains challenging to quantify void spatial distributions, especially in realistic granular packing with irregular grain shapes. This paper presents a three-dimensional (3D) framework for Voronoi analysis of realistic grain packing, based on a combination of modern X-ray computed tomography (XCT), quantitative image processing and computer simulation using the discrete element method (DEM). We also introduce an efficient and robust parallel processing tool, PySVT, based on Set Voronoi tessellation (SVT). 3D reconstruction of a realistic Ottawa sand assembly is conducted, and two numerical packings of ellipsoidal and spherical particles are generated to reproduce the structure of the realistic packing with the same grain size distribution and global void ratio (or porosity). A further comparison is made for a class of microstructural properties among these packings. For the solid (particle) phase, analyses of particle shape characteristics and contact network indicate statistically significant deviation occurs between 2D and 3D particle characteristics, in particular sphericity and roundness. The analyses for the void phase place an emphasis on the morphology of Voronoi cells and the local porosity. A log-normal distribution is found to describe well both the local porosity and the reduced surface area of Voronoi cells, which is in agreement with observations for numerical packings of non-realistic grains.
What problem does this paper attempt to address?