ARTIFICIAL NEURAL NETWORK FOR MODELING OF CU(II) BIO-SORPTION FROM SIMULATED WASTEWATER BY FUNGAL BIOMASS

Abstract

A three-layer artificial neural network model was developed to predict the removal efficiency of Cu(II) ions from simulated wastewater by fungal biomass based on 85 batch experiments. The effect of different parameters such as contact time between adsorbate and adsorbent (10-180 min), initial pH of the solution (3-7), initial metal concentration (50-250 mg/L), adsorbent dosage (0.05-2 g/100 mL), agitation speed (0-250 rpm) and temperature (10-60 ºC) were studied. The best values of these parameters that achieved the maximum removal efficiency (=95 %) of Cu(II) were 90 min, 6, 50 mg/L, 2 g/100 mL, 200 rpm, and 20 ºC, respectively. The present model was able to predict adsorption efficiency with a tangent sigmoid transfer function (tansig) at the hidden layer with 8 neurons and a linear transfer function (purelin) at the output layer. The linear regression between the network outputs and the corresponding targets was proven to be satisfactory with a correlation coefficient of greater than 0.99778 for the used six model variables. The sensitivity analysis based on the artificial neural network indicated that the initial pH of the solution with a relative importance of 22.1% appeared to be the most influential parameter in the Cu(II) removal, followed by dosage (19.5%), agitation speed (18.2%), temperature (14.1%), time (13.3%), and concentration (12.8%).