Abstract: In this study a ternary system containing sodium
chloride as solute, water as primary solvent and ethanol as the
antisolvent was considered to investigate the application of artificial
neural network (ANN) in prediction of sodium solubility in the
mixture of water as the solvent and ethanol as the antisolvent. The
system was previously studied using by Extended UNIQUAC model
by the authors of this study. The comparison between the results of
the two models shows an excellent agreement between them
(R2=0.99), and also approves the capability of ANN to predict the
thermodynamic behavior of ternary electrolyte systems which are
difficult to model.
Abstract: The recommended limit for cadmium concentration in
potable water is less than 0.005 mg/L. A continuous biosorption
process using indigenous red seaweed, Gracilaria corticata, was
performed to remove cadmium from the potable water. The process
was conducted under fixed conditions and the breakthrough curves
were achieved for three consecutive sorption-desorption cycles. A
modeling based on Artificial Neural Network (ANN) was employed
to fit the experimental breakthrough data. In addition, a simplified
semi empirical model, Thomas, was employed for this purpose. It
was found that ANN well described the experimental data (R2>0.99)
while the Thomas prediction were a bit less successful with R2>0.97.
The adjusted design parameters using the nonlinear form of Thomas
model was in a good agreement with the experimentally obtained
ones. The results approve the capability of ANN to predict the
cadmium concentration in potable water.