Product Distributions of Benzene Alkylation with Propylene Estimation Using Artificial Neural Network (ANN)

Abstract

The usability of artificial neural networks (ANN) for the estimation of the product distributions was investigated. The study was performed by following a model catalysis reaction, benzene alkylation with propylene on zeolite MCM-22. The effects of temperature, the ratio of benzene/propylene (B/P) and weight hourly space velocity (WHSV) on the product distributions were studied. Data obtained from different courses were used for training of the ANN and one set of data obtained from another courses was used for testing of the trained network. This network was designed as a Back-Propagation (BP) network with four neurons in the input layer, N neurons in the hidden layer and one neuron in the output layer. The network was trained till the mean square value between the targets and the outputs obtained was 1×10-4. The product distributions for the isopropylbenzene, di-isopropylbenzene and tri-isopropylbenzene were estimated using the trained network. The regression coefficient of determination showed a good correlation between estimated and experimental data sets for both train and test data sets. There are high correlations between experimental and estimated time course curves and that was another proof of the high performance of ANN for estimation of the product distributions of alkylation reaction.