Support vector machine-based modeling of grafting hyperbranched polyethylene glycol on polyethersulfone ultrafiltration membrane for separation of oil–water emulsion

Abstract

In present study, a new method based on the support vector machine (SVM) approach was employed to calculate the oil–water permeation flux and grafting yield of maleic anhydride and hyperbranched polyethylene glycol (PEG) on the polyethersulfone (PES) membrane surface. A set of 7 input/output experimental data was applied for training and testing the results of the model. The results of the developed SVM model showed good compliance with the experimental data, as the value of correlation index for the developed model is 0.96, 0.95 and 0.98 for the maleic anhydride and hyperbranched PEG grafting and oil–water permeation flux, respectively. The model was also validated against the kinetic reaction of benzophenone formation on the PES membrane which derives from combination of benzophenone formation reactions and the Lambert–Beer law. Also, to validate the SVM results, the oil–water permeation flux and antifouling characteristics of the modified and neat membranes were investigated, and based on the results, the SVM can be considered as a strong modeling approach in membrane process technologies.