Development of an artificial neural network correlation for prediction of overall gas holdup in bubble column reactors

In the literature, several correlations have been proposed for gas holdup prediction in bubble columns. However, these correlations fail to predict gas holdup over a wide range of conditions. Based on a databank of around 3500 measurements collected from the open literature, a correlation for gas holdup was derived using a combination of Dimensional Analysis and artificial neural network (ANN) modeling. The overall gas holdup was found to be a function of four dimensionless groups: Re-g, Fr-g, Eo/Mo, and rho(g)/rho(L). Statistical analysis showed that the proposed correlation has an average absolute relative error (AARE) of 15% and a standard deviation of 14%. A comparison with selected correlations in the literature showed that the developed ANN correlation noticeably improved prediction of overall gas holdup. The developed correlation also shows better prediction over a wide range of operating conditions, physical properties, and column diameters, and it predicts properly the trend of the effect of the operating and design parameters on overall gas holdup. (C) 2003 Elsevier Science B.V. All rights reserved.