Artificial neural network approach towards the separation of ethylparaben and diclofenac using pseudo-emulsion hollow fiber strip dispersion technique

Abstract

Pseudo-emulsion hollow fiber strip dispersion technique is known to be an effective way to separate pollutants from industrial wastewater. In the present study, data driven model like artificial neural network was developed for the prediction of extraction of ethylparaben and diclofenac using pseudo-emulsion hollow fiber strip dispersion technique. The feed, carrier and stripping phase concentration were taken as input parameters, whereas percentage of the extraction was chosen as an output parameter. The models were developed by carrying out the statistical analysis of parameters namely; root mean square error and mean absolute percentage error. The regression values achieved for training data set were 0.9956 and 0.97562 for ethylparaben and diclofenac separation, respectively. The results demonstrated that the artificial neural network model gives an accurate prediction of extraction data and hence can be quite helpful in designing the wastewater treatment plants.