Exploring the potential of transfer learning in extrapolating accelerated corrosion test data for long-term atmospheric corrosion forecasting

Vincent Vangrunderbeek; Leonardo Bertolucci Coelho; Dawei Zhang; Yiran Li; Yves Van Ingelgem; Herman Terryn

doi:https://doi.org/10.1016/j.corsci.2023.111619

Exploring the potential of transfer learning in extrapolating accelerated corrosion test data for long-term atmospheric corrosion forecasting

Vincent Vangrunderbeek, Leonardo Bertolucci Coelho, Dawei Zhang, Yiran Li, Yves Van Ingelgem, Herman Terryn

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

9 Citations (Scopus)

Abstract

This study utilizes transfer learning (TL) to enhance long-term atmospheric corrosion predictions. Using a Fe/Cu galvanic-type sensor, we gathered data in a controlled SAE J2334 salt spray setting and transferred this to an uncontrolled outdoor environment. Among TL methods tested, freezing the initial layer and fine-tuning others at a lower rate was most effective. The approach excelled at forecasting outdoor corrosion behaviour using a limited dataset. This approach could provide a solution to extrapolate results from controlled corrosion tests to unpredictable outdoor conditions and addressing data scarcity in machine learning modelling in the context of atmospheric corrosion.

Original language	English
Article number	111619
Number of pages	11
Journal	Corrosion Science
Volume	225
DOIs	https://doi.org/10.1016/j.corsci.2023.111619 https://doi.org/10.1016/j.corsci.2023.111619
Publication status	Published - Dec 2023

Bibliographical note

Funding Information:
The author L.B. Coelho is a Postdoctoral Researcher at the Fonds de la Recherche Scientifique – FNRS (Belgium) which is gratefully acknowledged. We would like to acknowledge that the SAE J2334 salt spray tests were conducted at the testing facilities of OCAS NV. We also wish to recognize the contributions of the National Materials Corrosion and Protection Data Center for supplying the sensor and the outdoor exposure data.". The author V. Vangrunderbeek would like to thank G. M. Paldino for fruitful discussions on data manipulation and analysis.

Funding Information:
The author L.B. Coelho is a Postdoctoral Researcher at the Fonds de la Recherche Scientifique – FNRS (Belgium) which is gratefully acknowledged. We would like to acknowledge that the SAE J2334 salt spray tests were conducted at the testing facilities of OCAS NV. We also wish to recognize the contributions of the National Materials Corrosion and Protection Data Center for supplying the sensor and the outdoor exposure data.". The author V. Vangrunderbeek would like to thank G. M. Paldino for fruitful discussions on data manipulation and analysis.

Publisher Copyright:
© 2023 Elsevier Ltd

Keywords

Atmospheric corrosion
Carbon steel
Weight loss

Access to Document

Cite this

@article{5e3e23053bf9428e9b932b36c9534d32,

title = "Exploring the potential of transfer learning in extrapolating accelerated corrosion test data for long-term atmospheric corrosion forecasting",

abstract = "This study utilizes transfer learning (TL) to enhance long-term atmospheric corrosion predictions. Using a Fe/Cu galvanic-type sensor, we gathered data in a controlled SAE J2334 salt spray setting and transferred this to an uncontrolled outdoor environment. Among TL methods tested, freezing the initial layer and fine-tuning others at a lower rate was most effective. The approach excelled at forecasting outdoor corrosion behaviour using a limited dataset. This approach could provide a solution to extrapolate results from controlled corrosion tests to unpredictable outdoor conditions and addressing data scarcity in machine learning modelling in the context of atmospheric corrosion.",

keywords = "Atmospheric corrosion, Carbon steel, Weight loss",

author = "Vincent Vangrunderbeek and Coelho, {Leonardo Bertolucci} and Dawei Zhang and Yiran Li and {Van Ingelgem}, Yves and Herman Terryn",

note = "Funding Information: The author L.B. Coelho is a Postdoctoral Researcher at the Fonds de la Recherche Scientifique – FNRS (Belgium) which is gratefully acknowledged. We would like to acknowledge that the SAE J2334 salt spray tests were conducted at the testing facilities of OCAS NV. We also wish to recognize the contributions of the National Materials Corrosion and Protection Data Center for supplying the sensor and the outdoor exposure data.{"}. The author V. Vangrunderbeek would like to thank G. M. Paldino for fruitful discussions on data manipulation and analysis. Funding Information: The author L.B. Coelho is a Postdoctoral Researcher at the Fonds de la Recherche Scientifique – FNRS (Belgium) which is gratefully acknowledged. We would like to acknowledge that the SAE J2334 salt spray tests were conducted at the testing facilities of OCAS NV. We also wish to recognize the contributions of the National Materials Corrosion and Protection Data Center for supplying the sensor and the outdoor exposure data.{"}. The author V. Vangrunderbeek would like to thank G. M. Paldino for fruitful discussions on data manipulation and analysis. Publisher Copyright: {\textcopyright} 2023 Elsevier Ltd",

year = "2023",

month = dec,

doi = "https://doi.org/10.1016/j.corsci.2023.111619",

language = "English",

volume = "225",

journal = "Corrosion Science",

issn = "0010-938X",

publisher = "Elsevier Limited",

}

Exploring the potential of transfer learning in extrapolating accelerated corrosion test data for long-term atmospheric corrosion forecasting. / Vangrunderbeek, Vincent ; Coelho, Leonardo Bertolucci; Zhang, Dawei; Li, Yiran; Van Ingelgem, Yves ; Terryn, Herman.

In: Corrosion Science, Vol. 225, 111619, 12.2023.

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

TY - JOUR

T1 - Exploring the potential of transfer learning in extrapolating accelerated corrosion test data for long-term atmospheric corrosion forecasting

AU - Vangrunderbeek, Vincent

AU - Coelho, Leonardo Bertolucci

AU - Zhang, Dawei

AU - Li, Yiran

AU - Van Ingelgem, Yves

AU - Terryn, Herman

N1 - Funding Information: The author L.B. Coelho is a Postdoctoral Researcher at the Fonds de la Recherche Scientifique – FNRS (Belgium) which is gratefully acknowledged. We would like to acknowledge that the SAE J2334 salt spray tests were conducted at the testing facilities of OCAS NV. We also wish to recognize the contributions of the National Materials Corrosion and Protection Data Center for supplying the sensor and the outdoor exposure data.". The author V. Vangrunderbeek would like to thank G. M. Paldino for fruitful discussions on data manipulation and analysis. Funding Information: The author L.B. Coelho is a Postdoctoral Researcher at the Fonds de la Recherche Scientifique – FNRS (Belgium) which is gratefully acknowledged. We would like to acknowledge that the SAE J2334 salt spray tests were conducted at the testing facilities of OCAS NV. We also wish to recognize the contributions of the National Materials Corrosion and Protection Data Center for supplying the sensor and the outdoor exposure data.". The author V. Vangrunderbeek would like to thank G. M. Paldino for fruitful discussions on data manipulation and analysis. Publisher Copyright: © 2023 Elsevier Ltd

PY - 2023/12

Y1 - 2023/12

N2 - This study utilizes transfer learning (TL) to enhance long-term atmospheric corrosion predictions. Using a Fe/Cu galvanic-type sensor, we gathered data in a controlled SAE J2334 salt spray setting and transferred this to an uncontrolled outdoor environment. Among TL methods tested, freezing the initial layer and fine-tuning others at a lower rate was most effective. The approach excelled at forecasting outdoor corrosion behaviour using a limited dataset. This approach could provide a solution to extrapolate results from controlled corrosion tests to unpredictable outdoor conditions and addressing data scarcity in machine learning modelling in the context of atmospheric corrosion.

AB - This study utilizes transfer learning (TL) to enhance long-term atmospheric corrosion predictions. Using a Fe/Cu galvanic-type sensor, we gathered data in a controlled SAE J2334 salt spray setting and transferred this to an uncontrolled outdoor environment. Among TL methods tested, freezing the initial layer and fine-tuning others at a lower rate was most effective. The approach excelled at forecasting outdoor corrosion behaviour using a limited dataset. This approach could provide a solution to extrapolate results from controlled corrosion tests to unpredictable outdoor conditions and addressing data scarcity in machine learning modelling in the context of atmospheric corrosion.

KW - Atmospheric corrosion

KW - Carbon steel

KW - Weight loss

UR - http://www.scopus.com/inward/record.url?scp=85175208853&partnerID=8YFLogxK

U2 - https://doi.org/10.1016/j.corsci.2023.111619

DO - https://doi.org/10.1016/j.corsci.2023.111619

M3 - Article

AN - SCOPUS:85175208853

VL - 225

JO - Corrosion Science

JF - Corrosion Science

SN - 0010-938X

M1 - 111619

ER -

Exploring the potential of transfer learning in extrapolating accelerated corrosion test data for long-term atmospheric corrosion forecasting

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this