Mass transfer analysis in ozone bubble columns using artificial neural networks

The design of ozone bubble columns is associated with accurate determination of some nonlinear parameters. The overall mass transfer coefficient (k_La) is the most important parameter as it dictates the efficiency of the bubble column. A multi-layer perceptron (MLP) artificial neural network (ANN) was used to simulate and predict the k_La in different ozone bubble columns by utilising simple inputs such as bubble column's geometry and operating conditions. The developed ANN model predicted k_La values in the training and validation data sets with a coefficient of multiple determination (R²) values that exceeded 0.87 and 0.85, respectively, which imply good model predictions.