EMMS-based Continuum Modeling Of Heterogeneous Gas-solid Flow In CFB Risers

Circulating fluidized bed (CFB) riser has been widely used in coal and biomass combustion, fluid catalytic cracking (FCC) and other modern chemical and energy processes. Due to the complexity of gas-solid flow and the existence of multiscale spatio-temporal heterogeneous structures, our understanding of the complex gas-solid flow is far from satisfactory. With increasing computing capacity of computers and improving numerical algorithm, more and more researchers are prone to study multiphase flow by means of computational fluid dynamics (CFD).Recently, continuum model (CM) is the most popular CFD method when studying industrial scale CFB risers. Due to the existence of multiscale spatio-temporal structures, it is necessary to take the effect of heterogeneous structures into account when simulating industrial-scale reactor by the means of CM. Nowadays Energy Minimization Multi-scale (EMMS) method is one of the most popular methods that are used to consider the effect of heterogeneous structures. Recent studies indicate that there are two ways to realize the EMMS-based continuum model:i) one is to use the EMMS drag model to modify the drag force in standard continuum model so that the effect of heterogeneous structures is considered, which is named as CM+EMMS method; ii) according to the analysis and validation of EMMS theory, it is more reasonable to assume the dilute phase and the dense phase as the two interpenetrating continua, which correspond to the gas-dominated mechanism and the particle-dominated mechanism respectively. Based on this understanding, an EMMS-based two-fluid model (EFM) is built, which can be regarded as the second way to realize the EMMS-based continuum model.In this dissertation, systematic studies of CM+EMMS (Chapter 2,3 and 4) and EFM (Chapter 5 and 6) were conducted. Chapter 2-4 includes the study of the effect of drag correlations in the EMMS drag model, the construction of type-A EMMS model and extended EMMS drag model for bidisperse gas-solid flow respectively. Chapter 5 and 6 focus on the built and validation of the EFM. Finally, these two approaches were compared with each other in the study of the effect of particle property and operating conditions in a pilot-scale circulating fluidized bed (Chapter 7). Several conclusions were carried out follows:(i) Although the output parameters (heterogeneous index) from the EMMS drag models were affected much by the choice of inputting drag correlations, CFD results using the EMMS drag model from different inputting drag correlations were almost identical and all in good agreement with experimental data.(ii) The difference between the simulation results obtained from the present model (type-A) and those obtained using previous model (type-B) were minor, thus confirming that the existing CM+EMMS method was feasible by transforming the effective type-B drag coefficient obtained from type-B EMMS drag model into type-A and then integrating into type-A CM.(iii) The extended EMMS drag model for bidisperse gas-solid flow could lead to reasonable mixing and segregation behaviors.(iv) This newly built EFM was validated by comparing the prediction of solid distribution in low-flux and high-flux riser with experimental data.(v) The prediction from CM+EMMS and EFM were in good agreement with experimental data and captured the innate heterogeneity in riser successfully, validating the feasibility of these EMMS-based methods.