Investigating and modeling the cleaning-in-place process for retrieving the membrane permeate flux: Case study of hydrophilic polyethersulfone (PES)

In this work the effects of backwash pressure, duration of acid and sodium hydroxide backwashing, sodium hydroxide concentration, and the duration of forward washing on performance of permeate flux recovery (PFR) were investigated. A two-level fractional factorial design (FFD) was used to design the experiments. The ability of back propagation neural network (BPNN) and radial basis function neural network (RBFNN) in predicting the performance of cleaning-in-place (CIP) of hydrophilic polyethersulfone (PES) membrane were investigated. It is found that BPNN has better ability in predicting the PFR performance than RBFNN. The best architecture of BPNN was a network consisting of 1 hidden layer with 15 neurons, which predicted the output values with a high determination coefficient (R-2) value of 0.9843. (C) 2016 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.