PLAYER

This project aims at understanding and modelling the behaviour of non spherical inertial particles transported by turbulent flows. The modelling of dispersed phase flows as classically developed relies on the assumption that the particles are spherical. The main objective of the present project is to include the non sphericity of particles in the modelling in order to investigate macroscopic numerical simulations of particulate flows with industrial CFD tools. This study will be devoted to the dispersion mechanisms (interaction between turbulence and particles), and the collisions and rebounds on the walls, which are the main physical phenomena governing the dynamics of the dispersed phase. In this way, it is proposed to benefit from the complementary competences of each partner (LEMTA, IMFT, I2M) to solve the scientific difficulty associated to the non sphericity of the particles with a particular attention: working out of the Stokes regime (low particle Reynolds numbers), for which many research works exist in the literature. However, carrying out such simulations at a macroscopic scale requires defining exactly the hydrodynamic forces and the couple exerted on the particles. Similarly, the rebounds of non spherical particles on the walls will be a key aspect that will be also considered. Indeed, a macroscopic simulation requires the introduction of boundary conditions for the dispersed phase. Few existing models are devoted to this behaviour near the wall. In this way, the three partners put their complementary competences into common to significantly increase the understanding of the physical phenomena, the representativity of the models and the efficiency of the numerical methods. The present work is naturally split in several complementary tasks and will began by numerical simulations led at the particle scale (microscopic scale, I2M) so as to access to exact data which will be shared between the partners. They will be introduced in the simulations at a mesoscopic scale for which the collective effect of an ensemble of particles will be studied by a direct numerical simulation DNS of the fluid flow coupled to a Lagrangian DPS modelling. The dispersion as well as the rebounds of the particles on the walls will be then studied under a stochastic modelling point of view (LEMTA, IMFT). To reach the objective of performing numerical simulations at a macroscopic scale, simulations will be run with dedicated CFD tools of each partner. All the cleverness of the results will in this way been put into stress.