Modeling on the Hydrodynamics of a High-Flux Circulating Fluidized Bed with Geldart Group A Particles by Kinetic Theory of Granular Flow

Abstract

A Eulerian multiphase model with the kinetic theory of granular flow (KTGF) was studied for modeling the hydrodynamic behaviors of high-flux circulating fluidized beds (HFCFBs) with Geldart group A particles. The sensitivities of several key models (i.e., turbulence model, drag model, and granular shear viscosity model) and modeling parameters (particle−particle restitution coefficient, particle−wall restitution coefficient, and specularity coefficient) on the predictions have been tested systematacially. Experimental results of Pärssinen and Zhu [AIChE J. 2001, 47 (10), 2197−2205; Chem. Eng. Sci. 2001, 56, 5295−5303] were used as a numerical benchmark to assess the simulations quantitatively. The results show that particle−particle and particle−wall restitution coefficients are not critical for the holistic distribution trends of solid volume fraction and solid velocity. A small specularity coefficient, such as φ = 0, could give the well predictions. The Syamlal−O’Brien drag model displays better agreement with individual radial distributions of both solid volume fraction and solid velocity, in terms of microscopic features. While the Syamlal shear viscosity model fails to obtain right trends at the upper parts of the riser. For turbulence model, the mixture turbulence model and per-phase turbulence model could not predict reasonable trend of radial solid velocity distribution along the riser. As a result, a group of suitable models and modeling parameters—i.e., dispersed turbulence model, Gidaspow shear viscosity model, Syamlal−O’Brien drag model, a specularity coefficient of φ = 0, a particle−particle restitution coefficient of e = 0.9, and a particle−wall restitution coefficient of ew = 0.99—are proposed for modeling Geldart group A particle flow in HFCFB risers. Finally, further validations have been conducted to confirm this suggestion, by comparing simulated results with experimental data.