Modelagem experimental e computacional DEM-CFD da fluidização de partículas Geldart-D com formato irregular

Abstract

Fluidized bed reactors find many applications in thermo-energy conversion processes. The preference for this type of system is due to the characteristics of high heat transfer and mass transfer rates between the particulate solid that composes the bed and the fluidizing agent. The three-dimensional computational simulation of these systems is able to provide local data such as velocity, pressure, temperature and chemical species concentration fields, which can not be obtained in zero-dimensional models or experimentally with the technology available today. Among the existing computational models, one of the most recent and attractive, due to its accuracy, is the one that uses the Discrete Element Method (DEM) and Computational Fluid Dynamics (CFD) coupling. In this approach, the motion of the granular phase is computed directly by the DEM, while the fluid is treated as a continuous medium by a CFD model. The interaction between the fluid and granular phases is performed by a coupling algorithm. With the objective of creating a DEM-CFD computational model for the fluidization of polypropylene particles, the physical characterization of the particulate material was carried out, determining the real and bulk mass densities, mean diameter, sphericity and repose and internal friction (drawdown) angles. The polypropylene particles were classified as of the Geldart-D type, and during their fluidization showed spout behavior. The particle-particle and particle-wall interaction coefficients, such as static friction, dynamic friction, restitution coefficient and rolling resistance, DEM input parameters were obtained through experiment planning and optimization technique. It was observed that among these parameters, what has the greatest effect on the results was the rolling resistance. Laboratory scale fluidizations provided reference values for minimum jet velocity, maximum bed pressure drop, and pressure drop in the minimum jet condition. During the simulations of the reactor, the refining of computational mesh was evaluated, as well as the refining of the wall. It was used four kinds of mesh, with the cells size based of particles diameters, for ranges of 2x, 3x and 5x, respectively. The DEM-CFD simulations of fluidization corroborate with the experimental results with an error of 1,88% to U_jm, 0,66% to 〖∆P〗_MÁX and 0,77% to 〖∆P〗_MÁX, being a powerful tool for studies of this species.