Neural Network Technique for Modeling of Cu (II) Removal from Aqueous Solution by Clinoptilolite

Clinoptilolite was investigated as an ion-exchange material for the removal of Cu (II) from aqueous solution. The effect of operational parameters such as pH, temperature, and initial concentration was studied to optimize the conditions for maximum removal of Cu (II) ions. Optimal operating conditions were determined to be a pH of 4, temperature of 30 A degrees C, and initial concentration of 0.361 mg/l to achieve equilibrium. After backpropagation (BP) training, the neural network (NN) techniques were able to predict the maximum removal efficiency with a tangent sigmoid transfer function (tansig) at hidden layer with 11 neurons and linear transfer function (purelin) at output layer. The Levenberg Marquardts backpropagation training algorithm was found as the best of 10 BP algorithms with mean-squared error of 0.000365. The results showed that NN techniques are able to predict the removal of Cu (II) from aqueous solution.