(474b) CFD Simulation of a Bubbling Fluidized Bed Using Reduced Gas-Particle Drag Based on Energy Minimization Multi-Scale (EMMS) Approach

Gas- particle flow systems have significant applications in chemical, pharmaceutical and energy industries. A computationally feasible approach to model gas-solid flows can be obtained from the averaged continuum equations of motion for both fluid and particles which is often called Two-Fluid Model (TFM). The averaging process leading to TFM equations, erases the details of flow at the level of individual particles; but their consequences appear in the averaged equations through terms for which one must develop interfacial relations (e.g. drag force). Two-fluid models for gas-solid flows reveal unstable heterogeneities whose length scales are as small as 10 particle diameters or smaller. Thus, if one wanted to solve the microscopic two-fluid model equations for gas-particle flow, grid sizes of the order of 10 particle diameters or smaller become essential. Such fine spatial resolution reduces the time steps required and results in significant increase in the computational time, especially for the industrial scale simulations. Homogeneous drag models (e.g. Syamlal O’Brien, Gidaspow and Wen & Yu) assume a homogeneous structure inside the control volumes, which is not valid, due to the aggregation of solid particles in the form of clusters. The effective drag coefficient in the actual systems will be lower than that in the homogeneous TFM to reflect the tendency of the gas to bypass cluster. The EMMS (Energy Minimization Multi-Scale) approach, which was used in this study for drag calculations, is based on the assumption that heterogeneous structures (i.e. clusters) with different sizes form and contribute to the drag reduction between gas and particulate phases. The resulting underdetermined set of equations is then solved by minimizing a function, called the stability condition. In this study an EMMS approach was derived and utilized for Geldart B particles used in CCS (Carbon Capture and Storage) units. Comparison of the bed expansion calculations based on EMMS with bed expansion calculations using homogeneous drag models reduced the deviation from experimental data from 70% to below 10%. In addition due to the fact that full EMMS approach requires significant computational times, a simplified EMMS approach, where Heterogeneity factor (H_d) is a only function of solid volume fraction was developed. The calculated bed height using the simplified EMMS compared well the bed height calculated using full EMMS while requiring considerably less computational time.