Update and Extension of the Mesoscale Model for Fast Fluidization: Toward Type C Choking, High-Density Fast-Bed, and Flow-Regime Map
Ming-Chuan Zhang
DOI: https://doi.org/10.1016/j.powtec.2023.118780
IF: 5.64
2023-01-01
Powder Technology
Abstract:As a supplementary paper for “Mesoscale analysis on clusters in conjunction with fast fluidized bed modeling” published recently, this article aims to update and further extend the model toward type C choking, high-density fast-bed (HDFF), and flow-regime map. Starting from an argument on mesoscale shear-layer formation, a more mechanistic expression for the energy recovery coefficient of in-cloud circulation was derived; and the effective velocity factor for gas flow in the dilute phase was also modified, with which the type C choking for Geldart Group B particles was correctly predicted. Two different mechanisms for “type C choking”, actually the transition at the upper limit of solids flux for traditional fast fluidization in the low velocity region, were identified, i.e. Choking C(I), defined by Gsmax at dMs+/dβ=dMs−/dβ; and Choking C(O), characterized by the cluster diameter reaching Dcl=dp. A rough estimation method for Choking C(I) or C(O) was obtained, thereby. Following the general strategy of extrapolation from its onset, the detailed procedures for modeling of the incipient stage of HDFF, and the classical case, were derived and demonstrated, respectively. The method to predict the upper limit of Gs for HDFF was also obtained, making the flow regime map for fast fluidization fully complete. Simple two parameter criteria, i.e. uf−ut/Dt0.29 and Gs/ρsuf−ut, for sketchy scale-up of fast-beds with the same gas-particle properties were proposed. From those, acceptable flow regime maps for enlarged risers were successfully estimated. Finally, the transition from dilute-transport to fast-bed was revisited, and the reasons for reduction of the solids-wall friction factor, from its direct regression result to the value used in the model, were discussed along with further outlook.