Dynamic modelling of crossflow microfiltration of bentonite suspension using recurrent neural networks

Using a set of experimental results (permeate flux and deposit thickness) a s a function of different operating conditions, obtained during crossflow m icrofiltration of a bentonite suspension with a laboratory pilot, a dynamic modelisation of this process by means of recurrent neural networks is prop osed. The elaborated neural network is able to describe the evolution of pe rmeate flux and deposit thickness from the process variables (transmembrane pressure, crossflow velocity, concentration of the suspension) and the sta rting point values for permeate flux and deposit thickness. The simulation of the evolution by such a model for experiments limited to a certain times pan, allows us to obtain coherent limit values for both permeate flux and d eposit thickness. (C) 1999 Elsevier Science S.A. All rights reserved.