There are lots of distinct advantages in bubbling fluidized beds(BFB), such as uniform temperature distribution, high mass and heat transfer rates and long residual time, so they have been used in a variety of industrial process. However, in most cases, the design of bubbling fluidized beds is still based on experience and empirical correlations which are rarely derived from underlying mechanisms, thus the predictability of these models is limited. In this paper, the BFB reactor is numerical simulated with TFM (Two Fluid Models) and FNI (Facet Nonuniformity Index) which derived from RNI (Radial Nonuniformity Index) to investigate the effect of four different structures inside BFB reactor, such as nozzle and air distributor. The kinetic theory of granular temperature was used to understand the collision mechanism among particles. The frictional stress term was added to the momentum equation to account for the long term and multi particle-particle contacts which avoided over-predicting the bed expansion. As usual, 2D geometry model can be chosen to save computational cost and get stable convergence, but it can not obtain full information to evaluate reactors due to ignoring the effect of velocity vector in one direction. Thus it will be more rigorous to predict real flow field in the bubbling fluidized beds with 3D model. The pressure drop of bed calculated by TFM was verified in experimental data. Correct pressure field can obtain the appropriate relationship of presser-velocity, then the velocity and concentration field were predicted from continuity and momentum equations with IPSA (Inter-Phase-Slip Algorithm). It was observed that bubble formation, growth and breakup have its own characteristics in each reactor due to effect of the different nozzles. The other properties, such as pressure fluctuation, air velocity and solid volume fraction, can also be obtained. To evaluate the performance of the BFB which have different internal structures quantitatively, FNI was introduced to describe the solid volume fraction distribution which can consider the impact of bubble size and bubble numbers. The proposed method above could help express detailed flow characteristics, which are used to design, scale-up and optimize reactors.
Key Words: bubbling fluidized bed, reactor design, numerical simulation, FNI