화학공학소재연구정보센터
Powder Technology, Vol.264, 423-429, 2014
Hydrodynamic studies on fluidized beds with internals: Experimental and ANN approach
In this investigation, hydrodynamic studies on fluidized beds with internals were conducted by varying the parameters such as inlet air flow rate, bed height, internal spacing, internal size and column diameter. Internals of width 25, 50 and 75% of the radius of the column were used. The effect of the above mentioned parameters on pressure drop, expansion ratio and fluctuation ratio was studied. The pressure drop was found to increase significantly with increase in bed height and decrease with increase in internal width. With internal, the pressure drop was seen to be only 30% of that with the unpromoted bed, thus resulting in lower power consumption. In the case of spacing, as the spacing was decreased the pressure drop was found to increase first and then decrease due to the effective breaking of bubbles in the presence of internals. Expansion ratio was found to reduce in the presence of internals and decrease with increase in internal width and with decrease in internal spacing. Fluctuation ratio was found to reduce in the presence of internals and decrease with increase in internal width and with decrease in internal spacing. The effect of internals on quality of fluidization was also studied. By the introduction of internals, the quality of fluidization in terms of fluidization index had improved. Correlations for pressure drop and expansion ratio have been developed with R-2 value of 0.998 and error of 0.076 for the pressure drop and R-2 value of 0.942 and error of 0.098 for the expansion ratio. A two layered feed forward back propagation Artificial Neural Network (ANN) model was developed to predict the pressure drop and expansion ratio and with 17 neurons, and it gave high R-2 values of 0.999 and 0.9898 for pressure drop and expansion ratio respectively with errors of 0.0148 and 0.0322. The model predictions and the experimental data are in excellent agreement. (C) 2014 Elsevier B.V. All rights reserved.