The GranOO workbench, a new tool for developing discrete element simulations, and its application to tribological problems

D Andre and JL Charles and I Iordanoff and J Neauport, ADVANCES IN ENGINEERING SOFTWARE, 74, 40-48 (2014).

DOI: 10.1016/j.advengsoft.2014.04.003

Discrete models are based on the descriptions of the physical states (e.g., velocity, position, temperature, magnetic momenta and electric potential) of a large number of discrete elements that form the media under study. These models are not based on a continuous description of the media. Thus, the models are particularly well adapted to describe the evolution of media driven by discontinuous phenomena such as multi- fracturation followed by debris flow as occurs in wear studies. Recently, the use of discrete models has been widened to face problems of complex rheological behaviors and/or multi-physical behaviors. Multi- physical problems involves complex mathematical formulations because of the combination of different families of differential equations when a continuous approach is chosen. These formulas are often much simpler to express in discrete models, in which each particle has a physical state and the evolution of that state is due to local physical interactions among particles. Since the year 2000, this method has been widely applied to the study of tribological problems including wear (Fillot et al., 2007) 1, the thermo-mechanical behavior of a contact (Richard et al., 2008) 2 and subsurface damage due to surface polishing (lordanoff et al., 2008) 3. Recent works have shown how this method can be used to obtain quantitative results (Andre et al., 2012) 4. To assist and promote research in this area, a free platform GranOO has been developed under a C++ environment and is distributed under a free GPL license. The primary features of this platform are presented in this paper. In addition, a series of examples that illustrate the main steps to construct a reliable tribological numerical simulation are detailed. The details of this platform can be found at (C) 2014 Elsevier Ltd. All rights reserved.

Return to Publications page