12

FIGURE 12.12 Comparison of simulation results with experimental data at low solids flux (dp 54 [xm, ps = 1545 kg m-3, D = 0.14 m, Ug = 4.33 ms-1, Gs = 10 kg m2 s-1) (from Ranade, 1999).

Dimensionlessladial£o-ordinate2

FIGURE 12.13 Sensitivity of simulation results to model parameters (dp = 54 [xm, ps = 1545 kg m-3, D = 0.14 m, Ug = 4.33 ms-1, Gs = 10 kg m-2s-1) (from Ranade, 1999).

Dimensionlessladial£o-ordinate2

FIGURE 12.13 Sensitivity of simulation results to model parameters (dp = 54 [xm, ps = 1545 kg m-3, D = 0.14 m, Ug = 4.33 ms-1, Gs = 10 kg m-2s-1) (from Ranade, 1999).

predicted results was also examined. Kuipers and coworkers (1991, 1998) used a value of speculiarity coefficient 0.5 while Bolio et al. (1995) used a very small value (0.002). The reduction in the value of speculiarity parameter causes flatter profiles of gas velocity (Fig. 12.12). An order of magnitude decrease in the value of speculiarity coefficient (0.05 from 0.5) increased the wall slip of solid particles from 0.9 to 3.2 m s-1. It can be seen that the predicted results obtained with the value 0.5 showed much better agreement with experimental data (Fig. 12.12). In view of these results, for all subsequent simulations, particle-particle restitution coefficient, particle-wall restitution coefficient and speculiarity coefficients were set to 1.0, 0.9 and 0.5, respectively. With these parameter settings, the computational model was found to give satisfactory agreement with the experimental data of Yang (1991).

To simulate gas-solid flows in industrial FCC risers, it is necessary to simulate flows at high solids fluxes. At higher solids flux, radial segregation increases and a significant downflow of solids may occur in the near-wall region in the riser. Several authors have reported such downflow of solids near the wall (van Breugel et al., 1969; Bader, 1988; Nieuwland et al, 1996; Derouin et al, 1997). We simulated the experimental conditions reported in these studies using the same computational model as was used to simulate the data of Yang (1991). Typical comparisons at higher solids fluxes are shown in Fig. 12.14. It can be seen that the agreement between predicted results and experimental data has significantly deteriorated and the model used in the present work failed to capture the significant downflow near the riser wall. It was interesting to note that the simulations showed the downflow at the wall if they were carried out without considering the gas phase turbulence model. Pita and Sundaresan (1991) showed reasonable agreement between predicted results and the experimental data of Bader et al. (1988) without including the turbulence model. Their

162 14*

I 62

"o

M 42

02 O.Ofl

0 0

Post a comment