Prediction of Gas Holdup in a Flotation Column by Artificial Neural Network

Abstract

Flotation column is one of the separation technologies investigated for removal of gangue mineral from coal fines. The effect of process variables, such as gas flow rate, feed flow rate, and frother dosage on gas holdup, were investigated. In this study, the measurement of gas holdup in presence of coal fines was carried out in a laboratory flotation column by a phase separation method. Detailed parametric study was performed to observe the effect of slurry concentration, gas flow rate, slurry flow rate, and frother concentration on gas holdup. An artificial neural network approach with three layers was systematically employed to predict gas holdup using some of the experimental results. The rest of the experimental results were successfully compared with the model predictions with less than 5% average error. In this work, emphasis was made on random selection of training data and small network. The developed trained network was also able to capture the non-linear prediction of gas holdup with new operating conditions that were not used in the training process and enhanced the physical understanding of the process.