Prediction of temperature profiles using artificial neural networks in a vertical thermosiphon reboiler

The objective of this work is to use artificial neural networks (ANNs) for predicting the temperature profiles as well as temperatures at various operating conditions in a vertical thermosiphon reboiler. The experimental data from the literature were used for training of feed forward artificial neural network with error back propagation technique. Gradient descent methods of optimization have been applied for training the network. It was observed that the predicted temperature profiles were very close to the actual experimental data. As the number of nodes increased, the training time decreased and the training was faster initially then it slowed down asymptotically. The prediction of ANN results was very close to the experimental values with a mean absolute relative error less than 4.3%.