Predicting RO/NF water quality by modified solution diffusion model and artificial neural networks

    Authors

    Y. Zhao; J. S. Taylor;S. Chellam

    Comments

    Authors: contact us about adding a copy of your work at STARS@ucf.edu

    Abbreviated Journal Title

    J. Membr. Sci.

    Keywords

    diffusion; water treatment; neural network; reverses osmosis; NANOFILTRATION; ULTRAFILTRATION; PERFORMANCE; PLANT; Engineering, Chemical; Polymer Science

    Abstract

    Membrane solute mass transfer is affected by physical-chemical properties of membrane films, solvent (water) and solutes. Existing mechanistic or empirical models that predict finished water quality from a diffusion controlled membrane can be significantly improved. Modelling membrane solute mass transfer by diffusion solution model is generally restricted to developing specific solute mass transfer coefficients that are site and stage specific. A modified solution diffusion model and two artificial neural network models have been developed for modelling diffusion controlled membrane mass transfer using stage specific solute MTCs. These models compensate for the effects of system flux, recovery and feed water quality on solute MTC and predict more accurately than existing models. (c) 2005 Elsevier B.V. All rights reserved.

    Journal Title

    Journal of Membrane Science

    Volume

    263

    Issue/Number

    1-2

    Publication Date

    1-1-2005

    Document Type

    Article

    Language

    English

    First Page

    38

    Last Page

    46

    WOS Identifier

    WOS:000232439300003

    ISSN

    0376-7388

    Share

    COinS