Neural networks simulation of the filtraton of sodium chloride and magnesium chloride solutions using nanofiltration membranes

N. A. Darwish, N. Hilal, H. Al-Zoubi, Abdul Wahab Mohammad

Research output: Contribution to journalArticle

33 Citations (Scopus)

Abstract

Artificial neural network (ANN) simulation is used to predict the rejection of two salts (NaCl and MgCl2) at typical seawater concentrations in a cross-flow nanofiltration membrane process. Rejections are predicted as a function of feed pressure and permeate flux with the salt concentration as a parameter. ANN predictions of the two salts' rejections are compared with experimental results obtained using three different nanofiltration membranes (NF90, NF270 and N30F) in a cross flow filtration set-up. The experimental program spans a concentration range of 5000 to 25 000 ppm for NaCl solution and of 5000 to 20 000 ppm for MgCl2 solution at pressures covering the range of 2-9 bars. The effects of the training algorithm, neural network architectures and transfer function on the ANN performance, as reflected by the percentage average absolute deviation are discussed. A network with one hidden layer comprising four neurones is found to be adequate for mapping input-output relationships and providing a good interpolative tool. Algorithms based on the conjugate gradient, regularization and quasi-Newton principles were found to outperform those based on the gradient descent and variable momentum gradient descent. The former algorithms gave an average absolute percentage deviation up to 5% whereas deviations up to 20% are obtained from the latter algorithms. For most of the cases considered, the ANN proved to be an adequate interpolation tool, where an excellent prediction was obtained for some concentration levels not represented in the training data set. For very limited cases, the ANN predictive performance was poor. However, the network's performance in these extreme cases didn't improve even with increasing the data points in the training data set.

Original languageEnglish
Pages (from-to)417-430
Number of pages14
JournalChemical Engineering Research and Design
Volume85
Issue number4 A
DOIs
Publication statusPublished - Apr 2007

Fingerprint

Nanofiltration membranes
Magnesium Chloride
Sodium chloride
Sodium Chloride
Magnesium
Neural networks
Salts
Network performance
Network architecture
Seawater
Neurons
Transfer functions
Interpolation
Momentum
Fluxes

Keywords

  • Artificial neuron networks
  • Membranes
  • Nanofiltration
  • Permeate flux
  • Rejection

ASJC Scopus subject areas

  • Polymers and Plastics

Cite this

Neural networks simulation of the filtraton of sodium chloride and magnesium chloride solutions using nanofiltration membranes. / Darwish, N. A.; Hilal, N.; Al-Zoubi, H.; Mohammad, Abdul Wahab.

In: Chemical Engineering Research and Design, Vol. 85, No. 4 A, 04.2007, p. 417-430.

Research output: Contribution to journalArticle

@article{2246cdda15f24386a67d87c071605cf9,
title = "Neural networks simulation of the filtraton of sodium chloride and magnesium chloride solutions using nanofiltration membranes",
abstract = "Artificial neural network (ANN) simulation is used to predict the rejection of two salts (NaCl and MgCl2) at typical seawater concentrations in a cross-flow nanofiltration membrane process. Rejections are predicted as a function of feed pressure and permeate flux with the salt concentration as a parameter. ANN predictions of the two salts' rejections are compared with experimental results obtained using three different nanofiltration membranes (NF90, NF270 and N30F) in a cross flow filtration set-up. The experimental program spans a concentration range of 5000 to 25 000 ppm for NaCl solution and of 5000 to 20 000 ppm for MgCl2 solution at pressures covering the range of 2-9 bars. The effects of the training algorithm, neural network architectures and transfer function on the ANN performance, as reflected by the percentage average absolute deviation are discussed. A network with one hidden layer comprising four neurones is found to be adequate for mapping input-output relationships and providing a good interpolative tool. Algorithms based on the conjugate gradient, regularization and quasi-Newton principles were found to outperform those based on the gradient descent and variable momentum gradient descent. The former algorithms gave an average absolute percentage deviation up to 5{\%} whereas deviations up to 20{\%} are obtained from the latter algorithms. For most of the cases considered, the ANN proved to be an adequate interpolation tool, where an excellent prediction was obtained for some concentration levels not represented in the training data set. For very limited cases, the ANN predictive performance was poor. However, the network's performance in these extreme cases didn't improve even with increasing the data points in the training data set.",
keywords = "Artificial neuron networks, Membranes, Nanofiltration, Permeate flux, Rejection",
author = "Darwish, {N. A.} and N. Hilal and H. Al-Zoubi and Mohammad, {Abdul Wahab}",
year = "2007",
month = "4",
doi = "10.1205/cherd06037",
language = "English",
volume = "85",
pages = "417--430",
journal = "Chemical Engineering Research and Design",
issn = "0263-8762",
publisher = "Institution of Chemical Engineers",
number = "4 A",

}

TY - JOUR

T1 - Neural networks simulation of the filtraton of sodium chloride and magnesium chloride solutions using nanofiltration membranes

AU - Darwish, N. A.

AU - Hilal, N.

AU - Al-Zoubi, H.

AU - Mohammad, Abdul Wahab

PY - 2007/4

Y1 - 2007/4

N2 - Artificial neural network (ANN) simulation is used to predict the rejection of two salts (NaCl and MgCl2) at typical seawater concentrations in a cross-flow nanofiltration membrane process. Rejections are predicted as a function of feed pressure and permeate flux with the salt concentration as a parameter. ANN predictions of the two salts' rejections are compared with experimental results obtained using three different nanofiltration membranes (NF90, NF270 and N30F) in a cross flow filtration set-up. The experimental program spans a concentration range of 5000 to 25 000 ppm for NaCl solution and of 5000 to 20 000 ppm for MgCl2 solution at pressures covering the range of 2-9 bars. The effects of the training algorithm, neural network architectures and transfer function on the ANN performance, as reflected by the percentage average absolute deviation are discussed. A network with one hidden layer comprising four neurones is found to be adequate for mapping input-output relationships and providing a good interpolative tool. Algorithms based on the conjugate gradient, regularization and quasi-Newton principles were found to outperform those based on the gradient descent and variable momentum gradient descent. The former algorithms gave an average absolute percentage deviation up to 5% whereas deviations up to 20% are obtained from the latter algorithms. For most of the cases considered, the ANN proved to be an adequate interpolation tool, where an excellent prediction was obtained for some concentration levels not represented in the training data set. For very limited cases, the ANN predictive performance was poor. However, the network's performance in these extreme cases didn't improve even with increasing the data points in the training data set.

AB - Artificial neural network (ANN) simulation is used to predict the rejection of two salts (NaCl and MgCl2) at typical seawater concentrations in a cross-flow nanofiltration membrane process. Rejections are predicted as a function of feed pressure and permeate flux with the salt concentration as a parameter. ANN predictions of the two salts' rejections are compared with experimental results obtained using three different nanofiltration membranes (NF90, NF270 and N30F) in a cross flow filtration set-up. The experimental program spans a concentration range of 5000 to 25 000 ppm for NaCl solution and of 5000 to 20 000 ppm for MgCl2 solution at pressures covering the range of 2-9 bars. The effects of the training algorithm, neural network architectures and transfer function on the ANN performance, as reflected by the percentage average absolute deviation are discussed. A network with one hidden layer comprising four neurones is found to be adequate for mapping input-output relationships and providing a good interpolative tool. Algorithms based on the conjugate gradient, regularization and quasi-Newton principles were found to outperform those based on the gradient descent and variable momentum gradient descent. The former algorithms gave an average absolute percentage deviation up to 5% whereas deviations up to 20% are obtained from the latter algorithms. For most of the cases considered, the ANN proved to be an adequate interpolation tool, where an excellent prediction was obtained for some concentration levels not represented in the training data set. For very limited cases, the ANN predictive performance was poor. However, the network's performance in these extreme cases didn't improve even with increasing the data points in the training data set.

KW - Artificial neuron networks

KW - Membranes

KW - Nanofiltration

KW - Permeate flux

KW - Rejection

UR - http://www.scopus.com/inward/record.url?scp=34247563914&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34247563914&partnerID=8YFLogxK

U2 - 10.1205/cherd06037

DO - 10.1205/cherd06037

M3 - Article

VL - 85

SP - 417

EP - 430

JO - Chemical Engineering Research and Design

JF - Chemical Engineering Research and Design

SN - 0263-8762

IS - 4 A

ER -