Predicting Mass Transfer in Pilot Scale External Loop Airlift Reactors using Neural Networks

Abstract

An artificial neural network (ANN) was built using selected input data of Newtonian fluids in pilot scale external loop airlift reactors of varying designs, in order to predict the mass transfer coefficient in other external loop airlift reactors with more general geometry. 663 data points were generated using air-glycerine and air-water systems in 5 different configurations of pilot scale external loop airlift reactors with 3 categories of sparger design. The data was modelled using the artificial neural network software, Predict (Version 3.30) by Neuralware. The correlation coefficient (R) for the neural network model was 0.98. The model was tested with unseen external data from various sources of which the R values ranged from 0.91 to 0.99. Additional external data, out of the experimental range of this investigation was evaluated, for which the R values ranged from 0.67 to 0.85. The ANN gave excellent approximations for the data within or below the training parameters.