Prediction of biosorption efficiency for the removal of copper(II) using artificial neural networks.

Various low-cost adsorbents have been used for removing Cu(II) ions from aqueous solutions for the treatment of copper containing wastewaters to remove organic compounds and color. Sawdust is an impressive adsorbent in terms of adsorption efficiency, cost and availability; hence the use of sawdust as biosorbent has been widely studied. Many earlier investigations tried to correlate the experimental data with available models or some modified empirical equations, but these results were unable to predict the values of parameters from a single equation. Artificial neural networks (ANN) are effective in modeling and simulation of highly non-liner multivariable relationships. A well-designed and very well trained network can converge even on multiple number of variables at a time without any complex modeling and empirical calculations. In this present work ANN is applied for the prediction of percentage adsorption efficiency for the removal of Cu(II) ions from aqueous solutions by sawdust. Artificial neural network model, based on multilayered partial recurrent back-propagation algorithm has been used. The performance of the network for predicting the sorption efficiency of sawdust for copper is found to be very impressive.