Enhancing efficiency of DEM modeling of particle breakage

Matteo Ciantia, Ningning Zhang, Marcos Arroyo

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In geotechnical practice, the discrete element method (DEM) is gaining wide acceptance as a powerful modelling tool. One field for which DEM is well adapted is that of crushable soils. Grain crushing has been modelled in DEM employing two alternative approaches: the multigenerational approach, in which single particles break and are replaced by new, smaller fragments; or by using agglomerates. The latter, despite being very helpful for the understanding of the micromechanics occurring in a single particle, becomes impractical when applied for modelling large scale problems. This work focuses on the enhancement of model efficiency from code-specific issues, as indicated in a series of simulation of high pressure isotropic compression of crushable sands. A recently developed model for crushable soils using multigenerational approach is adopted. It is shown that the advantageous code implementation adopted allows a considerable savings in computational cost with little influence on the accuracy in terms of grain size distribution evolution and mechanical behaviour.
Original languageEnglish
Title of host publicationProceedings of the 25th UKACM Conference on Computational Mechanics
EditorsAsaad Faramarzi, Samir Dirar
Place of PublicationBirmingham, United Kingdom
PublisherUniversity of Birmingham
Pages195-198
Number of pages4
Publication statusPublished - 2017
Event25th UKACM Conference on Computational Mechanics - University of Birmingham, Birmingham, United Kingdom
Duration: 11 Apr 201713 Apr 2017
http://ukacm2017.ukacm.org/

Conference

Conference25th UKACM Conference on Computational Mechanics
Country/TerritoryUnited Kingdom
CityBirmingham
Period11/04/1713/04/17
Internet address

Keywords

  • DEM
  • Grain Crushing
  • User defined model
  • Efficiency

Fingerprint

Dive into the research topics of 'Enhancing efficiency of DEM modeling of particle breakage'. Together they form a unique fingerprint.

Cite this