Multi-scale modelling of granular avalanches - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2013

Multi-scale modelling of granular avalanches

Résumé

Avalanches, debris flows, and landslides are geophysical hazards, which involve rapid mass movement of granular solids, water and air as a single-phase system. The dynamics of a granular flow involve at least three distinct scales: the micro-scale, meso-scale, and the macro-scale. This study aims to understand the ability of continuum models to capture the micro-mechanics of dry granular collapse. Material Point Method (MPM), a hybrid Lagrangian and Eulerian approach, with Mohr-Coulomb failure criterion is used to describe the continuum behaviour of granular column collapse, while the micromechanics is captured using Discrete Element Method (DEM) with tangential contact force model. The run-out profile predicted by the continuum simulations matches with DEM simulations for columns with small aspect ratios (‘h/r’ < 2), however MPM predicts larger run-out distances for columns with higher aspect ratios (‘h/r’ > 2). Energy evolution studies in DEM simulations reveal higher collisional dissipation in the initial free-fall regime for tall columns. The lack of a collisional energy dissipation mechanism in MPM simulations results in larger run-out distances. Micro-structural effects, such as shear band formations, were observed both in DEM and MPM simulations. A sliding flow regime is observed above the distinct passive zone at the core of the column. Velocity profiles obtained from both the scales are compared to understand the reason for a slow flow run-out mobilization in MPM simulations.

Dates et versions

hal-01567217 , version 1 (21-07-2017)

Licence

Copyright (Tous droits réservés)

Identifiants

Citer

Krishna Kumar, Kenichi Soga, Jean-Yves Delenne. Multi-scale modelling of granular avalanches. 7. International Conference on Micromechanics of Granular Media (Powders & Grains), Jul 2013, Sydney, Australia. ⟨10.1063/1.4812165⟩. ⟨hal-01567217⟩
66 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More