A comprehensive experimental and artificial network investigation of the performance of an ultrafiltration titanium dioxide ceramic membrane: application in produced water treatment

This work is an experimental investigation of the effects of the operating conditions on the performance of a novel titanium‐based ceramic ultrafiltration membrane used to treat field produced water. To design the experiments and optimize the operating conditions, the Taguchi method was used to predict the optimal operating conditions. Under optimal conditions, an almost oil free permeation is obtained (98.93%) along with the removal of more than 99% for the total organic carbon (TOC), a high turbidity removal (99.82%) and a good salinity rejection for a UF membrane. The membrane was capable of treating a high steady flux of 441 L/m 2 h with an overall flux decay of 28.6%. The Hermia’s cake formation model fitted the flux declining behaviour better than the three other associated models. Finally, four different techniques based on artificial intelligence (AI) methods were used to fit the flux declining behaviour. They seem to outperform the simple Hermiaˊs model for the modelling of oily water filtration.