Modeling the environmental dependence of pit growth using neural network approaches

M. K. Cavanaugh; R. G. Buchheit; N. Birbilis

doi:10.1016/j.corsci.2010.05.027

Modeling the environmental dependence of pit growth using neural network approaches

M. K. Cavanaugh
, R. G. Buchheit
, N. Birbilis

Research output: Contribution to journal › Article › peer-review

88 Scopus citations

Abstract

Corrosion pits have been shown to nucleate fatigue cracks, and this is a critical issue for aerospace aluminum alloys, which experience a variety of corrosive environments in service. Consequently, modeling pit growth as a function of environment is necessary. In this study, two orientations of AA7075-T651 blocks were boldly exposed in solutions of varying temperature, pH, and [Cl^-] for three exposure times. Optical profilometry and Weibull functions were utilized to characterize pit depth and diameter distributions. Artificial neural networks were a powerful tool in effectively modeling maximum pit dimensions and Weibull parameters. In most environments, pit growth followed t^1/3 kinetics.

Original language	English
Pages (from-to)	3070-3077
Number of pages	8
Journal	Corrosion Science
Volume	52
Issue number	9
DOIs	https://doi.org/10.1016/j.corsci.2010.05.027
State	Published - Sep 2010

Bibliographical note

Funding Information:
The authors gratefully acknowledge the support and financial assistance of the American Society for Engineering Education’s National Defense Science and Engineering Graduate Fellowship. Aspects of this work are sponsored by Defense Advanced Research Projects Agency under contract HR0111-04-C-0003, Northrop Grumman serving as the prime contractor. Additionally, the authors gratefully acknowledge the state government of Victoria (DIIRD) for funding of the Victorian Facility for Light Metals Surface Technology.

Funding

The authors gratefully acknowledge the support and financial assistance of the American Society for Engineering Education’s National Defense Science and Engineering Graduate Fellowship. Aspects of this work are sponsored by Defense Advanced Research Projects Agency under contract HR0111-04-C-0003, Northrop Grumman serving as the prime contractor. Additionally, the authors gratefully acknowledge the state government of Victoria (DIIRD) for funding of the Victorian Facility for Light Metals Surface Technology.

Funders	Funder number
American Society for Engineering Education’s National Defense Science and Engineering
DIIRD
Victorian Facility for Light Metals Surface Technology
Defense Advanced Research Projects Agency	HR0111-04-C-0003
State Government of Victoria

Keywords

A. Aluminum alloys
C. Neural networks
C. Pitting corrosion

ASJC Scopus subject areas

General Chemistry
General Chemical Engineering
General Materials Science

Access to Document

10.1016/j.corsci.2010.05.027

Cite this

@article{b1ff5f1e751b40c382e2554826835612,

title = "Modeling the environmental dependence of pit growth using neural network approaches",

abstract = "Corrosion pits have been shown to nucleate fatigue cracks, and this is a critical issue for aerospace aluminum alloys, which experience a variety of corrosive environments in service. Consequently, modeling pit growth as a function of environment is necessary. In this study, two orientations of AA7075-T651 blocks were boldly exposed in solutions of varying temperature, pH, and [Cl-] for three exposure times. Optical profilometry and Weibull functions were utilized to characterize pit depth and diameter distributions. Artificial neural networks were a powerful tool in effectively modeling maximum pit dimensions and Weibull parameters. In most environments, pit growth followed t1/3 kinetics.",

keywords = "A. Aluminum alloys, C. Neural networks, C. Pitting corrosion",

author = "Cavanaugh, \{M. K.\} and Buchheit, \{R. G.\} and N. Birbilis",

note = "Funding Information: The authors gratefully acknowledge the support and financial assistance of the American Society for Engineering Education{\textquoteright}s National Defense Science and Engineering Graduate Fellowship. Aspects of this work are sponsored by Defense Advanced Research Projects Agency under contract HR0111-04-C-0003, Northrop Grumman serving as the prime contractor. Additionally, the authors gratefully acknowledge the state government of Victoria (DIIRD) for funding of the Victorian Facility for Light Metals Surface Technology.",

year = "2010",

month = sep,

doi = "10.1016/j.corsci.2010.05.027",

language = "English",

volume = "52",

pages = "3070--3077",

number = "9",

}

TY - JOUR

T1 - Modeling the environmental dependence of pit growth using neural network approaches

AU - Cavanaugh, M. K.

AU - Buchheit, R. G.

AU - Birbilis, N.

N1 - Funding Information: The authors gratefully acknowledge the support and financial assistance of the American Society for Engineering Education’s National Defense Science and Engineering Graduate Fellowship. Aspects of this work are sponsored by Defense Advanced Research Projects Agency under contract HR0111-04-C-0003, Northrop Grumman serving as the prime contractor. Additionally, the authors gratefully acknowledge the state government of Victoria (DIIRD) for funding of the Victorian Facility for Light Metals Surface Technology.

PY - 2010/9

Y1 - 2010/9

N2 - Corrosion pits have been shown to nucleate fatigue cracks, and this is a critical issue for aerospace aluminum alloys, which experience a variety of corrosive environments in service. Consequently, modeling pit growth as a function of environment is necessary. In this study, two orientations of AA7075-T651 blocks were boldly exposed in solutions of varying temperature, pH, and [Cl-] for three exposure times. Optical profilometry and Weibull functions were utilized to characterize pit depth and diameter distributions. Artificial neural networks were a powerful tool in effectively modeling maximum pit dimensions and Weibull parameters. In most environments, pit growth followed t1/3 kinetics.

AB - Corrosion pits have been shown to nucleate fatigue cracks, and this is a critical issue for aerospace aluminum alloys, which experience a variety of corrosive environments in service. Consequently, modeling pit growth as a function of environment is necessary. In this study, two orientations of AA7075-T651 blocks were boldly exposed in solutions of varying temperature, pH, and [Cl-] for three exposure times. Optical profilometry and Weibull functions were utilized to characterize pit depth and diameter distributions. Artificial neural networks were a powerful tool in effectively modeling maximum pit dimensions and Weibull parameters. In most environments, pit growth followed t1/3 kinetics.

KW - A. Aluminum alloys

KW - C. Neural networks

KW - C. Pitting corrosion

UR - https://www.scopus.com/pages/publications/77953960106

UR - https://www.scopus.com/pages/publications/77953960106#tab=citedBy

U2 - 10.1016/j.corsci.2010.05.027

DO - 10.1016/j.corsci.2010.05.027

M3 - Article

AN - SCOPUS:77953960106

SN - 0010-938X

VL - 52

SP - 3070

EP - 3077

JO - Corrosion Science

JF - Corrosion Science

IS - 9

ER -

Modeling the environmental dependence of pit growth using neural network approaches

Abstract

Bibliographical note

Funding

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this