Numerical study of circulating fluidized beds built using Glicksman's simplified and full sets of scaling parameters

The objective of this study is to verify the hydrodynamic similarity between two scales of a circulating fluidized bed (CFB) reactor using computational fluid dynamics. Glicksman's full and simplified sets of scaling parameters were used to build two 1:4 reduced models for a CFB reactor. The model was validated against experimental results of the 3rd NETL/PSRI Challenge for fluidized beds. To exploit the hydrodynamic similarity between the two scales, dimensionless vertical pressure drop profile, horizontal profile of dimensionless solids vertical velocity and dimensionless solids density profiles were analyzed. Some perturbations were imposed with the objective of verifying the stability of the scaled models. Results showed that the simplified set scaled CFB presented a good hydrodynamic similarity with the real-scale reactor, but inferior to that of the full set scaled CFB. The perturbations implied in most cases on an increase of the mean relative error of hydrodynamic variables with respect to the CFB built with complete correspondence with the simplified set of scaling laws. The full set is a better choice in numerical simulations. However, in experimental studies, the simplified set is still a good choice due to its stability and its capability of keeping hydrodynamic similarity.