Using of Artificial Neural Network for Modeling of Isopropanol Dehydration by Pervaporation

Abstract

In this study, the isolation of water-isopropyl alcohol mixture was investigated using plate membrane (PERVAP 2210, Sulzer Chemtech) with the help of artificial neural network. Due to the fact that the water-isopropyl alcohol mixture contains azeotropes, it is difficult to separate this mixture. In this research, the results obtained from experimental tests are compared using an artificial neural network model. In this research, multilayer neural network has been used. About 70% of the available data is used for learning. Also, about 15% of the data is used for validation and the rest of the data for testing. The results obtained from neural network modeling showed good accuracy. The results showed that there was little error in predicting the experimental data.