Prediction of pressure drop and minimum spouting velocity in draft tube conical spouted beds using genetic programming approach

Abstract

The smart method of genetic programming (GP) is used to predict the operating pressure drop (ΔPs) and the minimum spouting velocity ums for conical spouted beds (CSBs) equipped with nonporous draft tubes. Accordingly, six dimensionless variables have been taken as model inputs, including crucial parameters associated with the bed and tube geometric and operating conditions. Two general correlations comprising almost all constitutive and operating variables have been derived for the first time by the GP approach. Both ΔPs and ums values predicted by the GP technique are in a fair agreement with the values corresponding to the experiments, with average absolute relative errors (AARE) of 18.9 and 19.9 %, respectively. The results of the proposed correlations show that the GP method is a powerful tool to make reasonable estimates.