Premium
Quantitative prediction of discrete element models on complex loading paths
Author(s) -
Sibille Luc,
Villard Pascal,
Darve Félix,
Aboul Hosn Rodaina
Publication year - 2019
Publication title -
international journal for numerical and analytical methods in geomechanics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.419
H-Index - 91
eISSN - 1096-9853
pISSN - 0363-9061
DOI - 10.1002/nag.2911
Subject(s) - constitutive equation , calibration , discrete element method , rotational symmetry , consistency (knowledge bases) , stress path , monotonic function , stress (linguistics) , spheres , stress space , finite element method , mathematics , perspective (graphical) , plane (geometry) , mechanics , structural engineering , mathematical analysis , geometry , physics , engineering , statistics , cauchy stress tensor , linguistics , philosophy , astronomy
Summary The ability of discrete element models to describe quantitatively (and not only qualitatively) the constitutive behaviour of a dense sand is assessed in this paper. Two kinds of 3D discrete models are considered. Both consider spheres as elementary particles. Nevertheless, the first model implements a contact law with rolling resistance whereas the second takes into account clumps made of two spheres. The discrete models are calibrated and validated from mechanical tests performed on a dense Hostun sand with a true triaxial apparatus. The calibration is carried out from axisymmetric drained compression tests, while the validation is discussed from monotonic and cyclic stress proportional loading paths and from a circular stress path in the deviatoric stress plane. The quality of the predictions of the discrete models are evaluated by comparison with the predictions given with advanced phenomenological constitutive relations, mainly an incrementally non‐linear relation. Predictions given by the discrete models are remarkable, particularly when it is put in perspective with respect to the very few number of mechanical tests required for their calibration. However, these results and conclusions were reached in enabling conditions, and some limitations of such discrete models should be kept in mind.