Estimating the Iodine number of activated carbon during thermal activation using Artificial Neural Networks (ANNs)

Artificial neural network, a biologically inspired computing method which has an ability to learn, self-adjust, and be trained, provides a powerful tool in solving pattern recognition problems. In this study, a new approach based on artificial neural networks (ANNs) has been designed to estimate the Iodine number of activated carbon prepared from Iranian pistachio shell using the thermal activation in special activation conditions. 75% of 108 experimental data of preparation of activated carbon from pistachio shell have been used to train the network. This data include Iodine adsorption capacity (Iodine number) versus temperature, activation time and oxidizing gas type. The present work, applied the Tan-sigmoid transfer function in two layers in the feedforward neural network with backpropagation algorithm. The results from the network are in good agreement with the experimental data and the maximum error is 0.015%. Finally, it is shown that the application of artificial neural networks in estimating the Iodine number of activated carbon prepared from pistachio shell can help us as a valuable tool to predict behavior of the activation in other conditions.

Estimating the Iodine number of activated carbon during thermal activation using Artificial Neural Networks (ANNs) Keywords:

Artificial neural networks , Feedforward Backpropagation , Iodine number , Activated carbon