A neural network prediction model of fluid displacements in porous media

Ali Elkamel, Mansour Karkoub, Ridha Gharbi

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

This paper presents the development and design of an artificial neural network that is able to predict the breakthrough oil recovery of immiscible displacement of oil by water in a two-dimensional vertical cross section. The data used in training the neural network was obtained from the results of fine-mesh numerical simulations. Several network architectures were investigated and trained using the back propagation with momentum algorithm. The neural network that gave the best predictive performance was a two-hidden layer network with 8 neurons in the first hidden layer and 8 neurons in the second hidden layer. This network also performed well against a cross validation test. The reservoir simulation data used so far in the training process was for a homogeneous reservoir, the more general case is still under investigation.

Original languageEnglish
JournalComputers and Chemical Engineering
Volume20
Issue numberSUPPL.1
Publication statusPublished - 1996
Externally publishedYes

Fingerprint

Porous materials
Neural networks
Neurons
Fluids
Oils
Network layers
Network architecture
Backpropagation
Momentum
Recovery
Water
Computer simulation

ASJC Scopus subject areas

  • Chemical Engineering(all)
  • Engineering(all)
  • Control and Systems Engineering

Cite this

A neural network prediction model of fluid displacements in porous media. / Elkamel, Ali; Karkoub, Mansour; Gharbi, Ridha.

In: Computers and Chemical Engineering, Vol. 20, No. SUPPL.1, 1996.

Research output: Contribution to journalArticle

@article{842edd9d348a49a4aebb48a595d2e468,
title = "A neural network prediction model of fluid displacements in porous media",
abstract = "This paper presents the development and design of an artificial neural network that is able to predict the breakthrough oil recovery of immiscible displacement of oil by water in a two-dimensional vertical cross section. The data used in training the neural network was obtained from the results of fine-mesh numerical simulations. Several network architectures were investigated and trained using the back propagation with momentum algorithm. The neural network that gave the best predictive performance was a two-hidden layer network with 8 neurons in the first hidden layer and 8 neurons in the second hidden layer. This network also performed well against a cross validation test. The reservoir simulation data used so far in the training process was for a homogeneous reservoir, the more general case is still under investigation.",
author = "Ali Elkamel and Mansour Karkoub and Ridha Gharbi",
year = "1996",
language = "English",
volume = "20",
journal = "Computers and Chemical Engineering",
issn = "0098-1354",
publisher = "Elsevier BV",
number = "SUPPL.1",

}

TY - JOUR

T1 - A neural network prediction model of fluid displacements in porous media

AU - Elkamel, Ali

AU - Karkoub, Mansour

AU - Gharbi, Ridha

PY - 1996

Y1 - 1996

N2 - This paper presents the development and design of an artificial neural network that is able to predict the breakthrough oil recovery of immiscible displacement of oil by water in a two-dimensional vertical cross section. The data used in training the neural network was obtained from the results of fine-mesh numerical simulations. Several network architectures were investigated and trained using the back propagation with momentum algorithm. The neural network that gave the best predictive performance was a two-hidden layer network with 8 neurons in the first hidden layer and 8 neurons in the second hidden layer. This network also performed well against a cross validation test. The reservoir simulation data used so far in the training process was for a homogeneous reservoir, the more general case is still under investigation.

AB - This paper presents the development and design of an artificial neural network that is able to predict the breakthrough oil recovery of immiscible displacement of oil by water in a two-dimensional vertical cross section. The data used in training the neural network was obtained from the results of fine-mesh numerical simulations. Several network architectures were investigated and trained using the back propagation with momentum algorithm. The neural network that gave the best predictive performance was a two-hidden layer network with 8 neurons in the first hidden layer and 8 neurons in the second hidden layer. This network also performed well against a cross validation test. The reservoir simulation data used so far in the training process was for a homogeneous reservoir, the more general case is still under investigation.

UR - http://www.scopus.com/inward/record.url?scp=0029656866&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0029656866&partnerID=8YFLogxK

M3 - Article

VL - 20

JO - Computers and Chemical Engineering

JF - Computers and Chemical Engineering

SN - 0098-1354

IS - SUPPL.1

ER -