CFD simulation of bubbling fluidized beds using a local-structure-dependent drag model

- The local-structure-dependent drag model based on a computational cell is established.
- A computational cell is resolved into three sub-systems.
- The meso-scale structural parameters and drag force are correlated by implementing the scale resolution.
- The grid independence of the new drag model is tested.
- The prediction accuracy of the new drag model is significantly improved compared with Gidaspow model.

Conventional drag models coupled with a two-fluid model (TFM) are subject to certain restrictions on the predictions of the hydrodynamic behaviors in bubbling fluidized beds with fine particles, owing to the lack of scale resolution and the subsequent neglect of the impact of the meso-scale structure on gas-solid interaction. In this work, a local control volume is resolved into three sub-systems (i.e., the emulsion phase, bubble phase, and interphase) by implementing a designed route of scale resolution. Then, a local-structure-dependent (LSD) drag model based on the energy-minimization multi-scale (EMMS) theory is developed to account for the dependence of the drag force on the meso-scale structure. The LSD drag model is solved using a genetic algorithm and the obtained heterogeneous drag forces are integrated with the TFM to simulate the hydrodynamic behaviors in bubbling fluidized beds for different gas-solid systems. Consequently, the proposed drag model is validated to provide satisfying predictions of the fluidizations of Geldart A, A/B, and B particles. Furthermore, bubble diameters obtained from computational fluid dynamics (CFD) simulations are compared with those obtained from the empirical or semi-empirical correlations. The results indicate that the LSD drag model can capture the gas-solid hydrodynamics in a bubbling fluidized bed.