Machine learning-based prediction of pitting corrosion resistance in stainless steels exposed to chloride environments

Chunyu Qiao; Hong Luo; Xuefei Wang; Hongxu Cheng; Da Bi; Xiaogang Li

doi:10.1016/j.colsurfa.2023.132274

Machine learning-based prediction of pitting corrosion resistance in stainless steels exposed to chloride environments

Chunyu Qiao
, Hong Luo^*
, Xuefei Wang
, Hongxu Cheng
, Da Bi
, Xiaogang Li

^*Corresponding author for this work

University of Science and Technology Beijing

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

21 Scopus citations

Abstract

Aiming at the limitation of the pitting resistance equivalent number theory in qualitative evaluation of pitting resistance, a new strategy based on extreme gradient regression (XGBR) is used to predict the E_pit of stainless steel (SS). The goodness of fit (R²) and root mean square error (RMSE) of the model on the test set were 0.877 and 132.51 mV, respectively. In addition, the generalization ability of the model is accepted on a validation set. This innovative strategy provides valuable insights to accurately predict the E_pit of stainless steel, which is helpful to design new SS with excellent pitting corrosion resistance.

Original language	English
Article number	132274
Journal	Colloids and Surfaces A: Physicochemical and Engineering Aspects
Volume	676
DOIs	https://doi.org/10.1016/j.colsurfa.2023.132274
State	Published - 5 Nov 2023
Externally published	Yes

Keywords

Machine learning
Pitting potential
PREN
Stainless steels
XGBR

Access to Document

10.1016/j.colsurfa.2023.132274

Cite this

@article{3e22924209764a20adb21bfb937cff4e,

title = "Machine learning-based prediction of pitting corrosion resistance in stainless steels exposed to chloride environments",

abstract = "Aiming at the limitation of the pitting resistance equivalent number theory in qualitative evaluation of pitting resistance, a new strategy based on extreme gradient regression (XGBR) is used to predict the Epit of stainless steel (SS). The goodness of fit (R2) and root mean square error (RMSE) of the model on the test set were 0.877 and 132.51 mV, respectively. In addition, the generalization ability of the model is accepted on a validation set. This innovative strategy provides valuable insights to accurately predict the Epit of stainless steel, which is helpful to design new SS with excellent pitting corrosion resistance.",

keywords = "Machine learning, Pitting potential, PREN, Stainless steels, XGBR",

author = "Chunyu Qiao and Hong Luo and Xuefei Wang and Hongxu Cheng and Da Bi and Xiaogang Li",

note = "Publisher Copyright: {\textcopyright} 2023 Elsevier B.V.",

year = "2023",

month = nov,

day = "5",

doi = "10.1016/j.colsurfa.2023.132274",

language = "英语",

volume = "676",

journal = "Colloids and Surfaces A: Physicochemical and Engineering Aspects",

issn = "0927-7757",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Machine learning-based prediction of pitting corrosion resistance in stainless steels exposed to chloride environments

AU - Qiao, Chunyu

AU - Luo, Hong

AU - Wang, Xuefei

AU - Cheng, Hongxu

AU - Bi, Da

AU - Li, Xiaogang

PY - 2023/11/5

Y1 - 2023/11/5

N2 - Aiming at the limitation of the pitting resistance equivalent number theory in qualitative evaluation of pitting resistance, a new strategy based on extreme gradient regression (XGBR) is used to predict the Epit of stainless steel (SS). The goodness of fit (R2) and root mean square error (RMSE) of the model on the test set were 0.877 and 132.51 mV, respectively. In addition, the generalization ability of the model is accepted on a validation set. This innovative strategy provides valuable insights to accurately predict the Epit of stainless steel, which is helpful to design new SS with excellent pitting corrosion resistance.

AB - Aiming at the limitation of the pitting resistance equivalent number theory in qualitative evaluation of pitting resistance, a new strategy based on extreme gradient regression (XGBR) is used to predict the Epit of stainless steel (SS). The goodness of fit (R2) and root mean square error (RMSE) of the model on the test set were 0.877 and 132.51 mV, respectively. In addition, the generalization ability of the model is accepted on a validation set. This innovative strategy provides valuable insights to accurately predict the Epit of stainless steel, which is helpful to design new SS with excellent pitting corrosion resistance.

KW - Machine learning

KW - Pitting potential

KW - PREN

KW - Stainless steels

KW - XGBR

UR - https://www.scopus.com/pages/publications/85168991206

U2 - 10.1016/j.colsurfa.2023.132274

DO - 10.1016/j.colsurfa.2023.132274

M3 - 文章

AN - SCOPUS:85168991206

SN - 0927-7757

VL - 676

JO - Colloids and Surfaces A: Physicochemical and Engineering Aspects

JF - Colloids and Surfaces A: Physicochemical and Engineering Aspects

M1 - 132274

ER -

Machine learning-based prediction of pitting corrosion resistance in stainless steels exposed to chloride environments

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this