چكيده به لاتين
Packed beds are widely used in industries and lots of studies have been done on the mass transfer, heat transfer, and pressure drop. The results of the extensive literature review showed that tortuosity and hot-spots have not been studied sufficiently. One of the main reason which makes it difficult to study these beds is the high ratio of bed length to its diameter and also the presence of large umber of catalysts inside them. Consequently, the mentioned parameters were studied in four parts. First, the effect of catalyst pellet configuration on the created hot-spots on the tube wall was studied. To this aim, the behavior of six non-spherical pellets (cylindrical, sg-cylindrical, tri-lobe, multi-hole cylindrical, multi-hole sg-cylindrical, multi-hole tri-lobe) in both horizontal and vertical positions close to the tube wall was studied. The results showed that the catalyst pellet configuration effect is considerable on the created hot-spots and the tri-lobe, sg-cylindrical, and cylindrical pellets are in priority in terms of reducing the hot-spots. In the second part, the Analysis of Variance (ANOVA) procedure was applied to find the optimal placement of the cylindrical pellet beside the wall. The heat transfer behavior of the cylindrical pellet around the X, Y, and Z direction was studied in 0-20, -45-45, and 0-30 degrees respectively and an equation was proposed to predict the maximum temperature on the tube wall. In the third part, the effect of tortuosity on the heat transfer and pressure drop of structured packed beds were studied. To this aim, the heat transfer and pressure drop of two sample beds with the same porosity and sphericity and different arrangements were studied. The streamlines data was extracted from simulation results and a self-modified code in MATLAB was used to calculate the bed tortuosity. It was observed that tortuosity effects have a great impact on the results and it can’t be predicted by existing correlations. Finally, the pressure drop and Nusselt number correlations were modified by adding tortuosity number which was calculated using the length of streamlines. In the fourth part, a procedure was proposed to simulate the long beds. The bed divided into four equal parts and the first section was simulated. The outlet of the first section was considered as the inlet boundary conditions of the second one and this procedure continued for remained sections. Finally, the results were compared with the main reactor and showed that the proposed procedure can be used instead of the simulation of the main bed (which is not possible/costly).
