Skip to main navigation Skip to search Skip to main content

Integrated characterization, classification, and quasi-3D reconstruction of highly irregular particles using multiscale shape descriptors for predictive DEM flow simulation

  • Aidana Boribayeva
  • , Sultan Sultaniyar
  • , Islam Lukmanov
  • , Assiya Baigarina
  • , Luis R. Rojas-Solórzano
  • , Jennifer Sinclair Curtis
  • , Nicolin Govender
  • , Boris Golman

Research output: Contribution to journalArticlepeer-review

Abstract

Accurate characterization and reconstruction of particle morphology are critical for generating realistic DEM inputs and achieving reliable predictions of granular flow in industrial processes. This study develops integrated workflow that performs particle shape classification and reconstruction to generate accurate DEM inputs and enhance the prediction of granular flow behavior. Highly irregular particles were imaged using a dynamic particle-tracking system, reducing orientation bias and enabling more representative shape distributions under realistic flow conditions. Particle outlines were analyzed using Fourier descriptors, and a harmonic order of k=16 identified as optimal for ensuring convergence of the calculations of elongation ratio, asymmetricity, polygonality, angularity, and surface roughness. These multiscale descriptors capture macro-, meso-, and micro-scale irregularities and serve as the basis for systematic particle shape classification into four morphological classes: Compact Angular, Compact Rounded, Elongated Angular, and Elongated Rounded. Reconstructed polyhedral particles preserved elongation, angularity, and shape complexity, providing accurate non-spherical particles for DEM input. The DEM simulations were performed for rolling–cascading flow in a rotating drum. The dynamic angle of repose was measured across three regions of the flowing bed, with minor deviations from experiments: the lower and upper regions within 1%–2% and the middle region slightly underestimated by 8.6%, highlighting the dominant role of particle geometry. Overall, this work integrates dynamic particle imaging, multiscale shape analysis, hierarchical classification, and quasi-3D reconstruction to establish a robust framework for enhancing DEM simulations of highly irregular particles with relevance to moving bed heat exchangers, rotary kilns, and bulk powder handling.

Original languageEnglish
Article number122435
JournalPowder Technology
Volume477
DOIs
Publication statusPublished - 15 Jun 2026
Externally publishedYes

Keywords

  • DEM calibration
  • Elliptical Fourier Analysis
  • Granular flow behavior
  • Irregular particles
  • Multiscale shape descriptors
  • Particle shape classification
  • Polyhedral models

ASJC Scopus subject areas

  • General Chemical Engineering

Fingerprint

Dive into the research topics of 'Integrated characterization, classification, and quasi-3D reconstruction of highly irregular particles using multiscale shape descriptors for predictive DEM flow simulation'. Together they form a unique fingerprint.

Cite this