Identification of a steady-state flow in porous media using artificial neural networks

Marek J. Lefik; Daniela P. Boso; Bernhard A. Schrefler

doi:10.1115/ESDA2012-82948

Identification of a steady-state flow in porous media using artificial neural networks

Marek J. Lefik, Daniela P. Boso, Bernhard A. Schrefler

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

For a steady state convection problem, assuming given concentration field values in a few measurement points and hydraulic head values in the same piezometers, the source of the concentration, and its intensity are deduced using Artificial Neural Networks (ANNs). ANNs are trained with data extracted from Finite Difference (FD) solution of a classical convection problem for small Peclet number. The numerical analysis is exemplified for vanishing, homogeneous and non-homogeneous field of velocity. It is shown that the diffusivity vector can also be identified. The complexity of the problem is discussed for each studied case.

Original language	English (US)
Title of host publication	ASME 2012 11th Biennial Conference on Engineering Systems Design and Analysis, ESDA 2012
Pages	89-95
Number of pages	7
Volume	1
DOIs	https://doi.org/10.1115/ESDA2012-82948
State	Published - Dec 1 2012
Event	ASME 2012 11th Biennial Conference on Engineering Systems Design and Analysis, ESDA 2012 - Nantes, France Duration: Jul 2 2012 → Jul 4 2012

Other

Other	ASME 2012 11th Biennial Conference on Engineering Systems Design and Analysis, ESDA 2012
Country/Territory	France
City	Nantes
Period	7/2/12 → 7/4/12

ASJC Scopus subject areas

Control and Systems Engineering
Mechanical Engineering

Access to Document

10.1115/ESDA2012-82948

Cite this

Lefik, MJ, Boso, DP & Schrefler, BA 2012, Identification of a steady-state flow in porous media using artificial neural networks. in ASME 2012 11th Biennial Conference on Engineering Systems Design and Analysis, ESDA 2012. vol. 1, pp. 89-95, ASME 2012 11th Biennial Conference on Engineering Systems Design and Analysis, ESDA 2012, Nantes, France, 7/2/12. https://doi.org/10.1115/ESDA2012-82948

@inproceedings{207d795ab971465bbbf9be7862a7f7ff,

title = "Identification of a steady-state flow in porous media using artificial neural networks",

abstract = "For a steady state convection problem, assuming given concentration field values in a few measurement points and hydraulic head values in the same piezometers, the source of the concentration, and its intensity are deduced using Artificial Neural Networks (ANNs). ANNs are trained with data extracted from Finite Difference (FD) solution of a classical convection problem for small Peclet number. The numerical analysis is exemplified for vanishing, homogeneous and non-homogeneous field of velocity. It is shown that the diffusivity vector can also be identified. The complexity of the problem is discussed for each studied case.",

author = "Lefik, {Marek J.} and Boso, {Daniela P.} and Schrefler, {Bernhard A.}",

year = "2012",

month = dec,

day = "1",

doi = "10.1115/ESDA2012-82948",

language = "English (US)",

isbn = "9780791844847",

volume = "1",

pages = "89--95",

booktitle = "ASME 2012 11th Biennial Conference on Engineering Systems Design and Analysis, ESDA 2012",

note = "ASME 2012 11th Biennial Conference on Engineering Systems Design and Analysis, ESDA 2012 ; Conference date: 02-07-2012 Through 04-07-2012",

}

TY - GEN

T1 - Identification of a steady-state flow in porous media using artificial neural networks

AU - Lefik, Marek J.

AU - Boso, Daniela P.

AU - Schrefler, Bernhard A.

PY - 2012/12/1

Y1 - 2012/12/1

N2 - For a steady state convection problem, assuming given concentration field values in a few measurement points and hydraulic head values in the same piezometers, the source of the concentration, and its intensity are deduced using Artificial Neural Networks (ANNs). ANNs are trained with data extracted from Finite Difference (FD) solution of a classical convection problem for small Peclet number. The numerical analysis is exemplified for vanishing, homogeneous and non-homogeneous field of velocity. It is shown that the diffusivity vector can also be identified. The complexity of the problem is discussed for each studied case.

AB - For a steady state convection problem, assuming given concentration field values in a few measurement points and hydraulic head values in the same piezometers, the source of the concentration, and its intensity are deduced using Artificial Neural Networks (ANNs). ANNs are trained with data extracted from Finite Difference (FD) solution of a classical convection problem for small Peclet number. The numerical analysis is exemplified for vanishing, homogeneous and non-homogeneous field of velocity. It is shown that the diffusivity vector can also be identified. The complexity of the problem is discussed for each studied case.

UR - http://www.scopus.com/inward/record.url?scp=84883856139&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84883856139&partnerID=8YFLogxK

U2 - 10.1115/ESDA2012-82948

DO - 10.1115/ESDA2012-82948

M3 - Conference contribution

AN - SCOPUS:84883856139

SN - 9780791844847

VL - 1

SP - 89

EP - 95

BT - ASME 2012 11th Biennial Conference on Engineering Systems Design and Analysis, ESDA 2012

T2 - ASME 2012 11th Biennial Conference on Engineering Systems Design and Analysis, ESDA 2012

Y2 - 2 July 2012 through 4 July 2012

ER -

Identification of a steady-state flow in porous media using artificial neural networks

Abstract

Other

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this