Artificial intelligence combined with high-throughput calculations to improve the corrosion resistance of AlMgZn alloy

Yucheng Ji; Xiaoqian Fu; Feng Ding; Yongtao Xu; Yang He; Min Ao; Fulai Xiao; Dihao Chen; Poulumi Dey; Wentao Qin; Kui Xiao; Jingli Ren; Decheng Kong; Xiaogang Li; Chaofang Dong

doi:10.1016/j.corsci.2024.112062

Artificial intelligence combined with high-throughput calculations to improve the corrosion resistance of AlMgZn alloy

Yucheng Ji, Xiaoqian Fu, Feng Ding, Yongtao Xu, Yang He, Min Ao, Fulai Xiao, Dihao Chen, Poulumi Dey, Wentao Qin, Kui Xiao, Jingli Ren, Decheng Kong, Xiaogang Li, Chaofang Dong^*

^*Corresponding author for this work

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

14 Citations (SciVal)

52 Downloads (Pure)

Abstract

Efficiently designing lightweight alloys with combined high corrosion resistance and mechanical properties remains an enduring topic in materials engineering. Due to the inadequate accuracy of conventional stress-strain machine learning (ML) models caused by corrosion factors, a novel reinforcement self-learning ML algorithm combined with calculated features (accuracy R² >0.92) is developed. Based on the ML models, calculated work functions and mechanical moduli, a Computation Designed Corrosion-Resistant Al alloy is fabricated and verified. The performance (elongation reaches ∼30 %) is attributed to the H trapping Al-Sc-Cu phases (-1.44 eV H⁻¹) and Cu-modified η/η' precipitates inside the grain boundaries (GBs).

Original language	English
Article number	112062
Number of pages	15
Journal	Corrosion Science
Volume	233
DOIs	https://doi.org/10.1016/j.corsci.2024.112062
Publication status	Published - 2024

Bibliographical note

Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care
Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

Keywords

Al-Zn-Mg alloys
First-principles calculation
Machine learning
Molecular dynamic simulation
Precipitates

Access to Document

10.1016/j.corsci.2024.112062

1-s2.0-S0010938X24002464-mainFinal published version, 17.8 MB

Cite this

@article{e02260c621344ac5b43a614e394eff1f,

title = "Artificial intelligence combined with high-throughput calculations to improve the corrosion resistance of AlMgZn alloy",

abstract = "Efficiently designing lightweight alloys with combined high corrosion resistance and mechanical properties remains an enduring topic in materials engineering. Due to the inadequate accuracy of conventional stress-strain machine learning (ML) models caused by corrosion factors, a novel reinforcement self-learning ML algorithm combined with calculated features (accuracy R2 >0.92) is developed. Based on the ML models, calculated work functions and mechanical moduli, a Computation Designed Corrosion-Resistant Al alloy is fabricated and verified. The performance (elongation reaches ∼30 %) is attributed to the H trapping Al-Sc-Cu phases (-1.44 eV H−1) and Cu-modified η/η' precipitates inside the grain boundaries (GBs).",

keywords = "Al-Zn-Mg alloys, First-principles calculation, Machine learning, Molecular dynamic simulation, Precipitates",

author = "Yucheng Ji and Xiaoqian Fu and Feng Ding and Yongtao Xu and Yang He and Min Ao and Fulai Xiao and Dihao Chen and Poulumi Dey and Wentao Qin and Kui Xiao and Jingli Ren and Decheng Kong and Xiaogang Li and Chaofang Dong",

note = "Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.",

year = "2024",

doi = "10.1016/j.corsci.2024.112062",

language = "English",

volume = "233",

journal = "Corrosion Science",

issn = "0010-938X",

publisher = "Elsevier",

}

TY - JOUR

T1 - Artificial intelligence combined with high-throughput calculations to improve the corrosion resistance of AlMgZn alloy

AU - Ji, Yucheng

AU - Fu, Xiaoqian

AU - Ding, Feng

AU - Xu, Yongtao

AU - He, Yang

AU - Ao, Min

AU - Xiao, Fulai

AU - Chen, Dihao

AU - Dey, Poulumi

AU - Qin, Wentao

AU - Xiao, Kui

AU - Ren, Jingli

AU - Kong, Decheng

AU - Li, Xiaogang

AU - Dong, Chaofang

N1 - Green Open Access added to TU Delft Institutional Repository 'You share, we take care!' - Taverne project https://www.openaccess.nl/en/you-share-we-take-care Otherwise as indicated in the copyright section: the publisher is the copyright holder of this work and the author uses the Dutch legislation to make this work public.

PY - 2024

Y1 - 2024

N2 - Efficiently designing lightweight alloys with combined high corrosion resistance and mechanical properties remains an enduring topic in materials engineering. Due to the inadequate accuracy of conventional stress-strain machine learning (ML) models caused by corrosion factors, a novel reinforcement self-learning ML algorithm combined with calculated features (accuracy R2 >0.92) is developed. Based on the ML models, calculated work functions and mechanical moduli, a Computation Designed Corrosion-Resistant Al alloy is fabricated and verified. The performance (elongation reaches ∼30 %) is attributed to the H trapping Al-Sc-Cu phases (-1.44 eV H−1) and Cu-modified η/η' precipitates inside the grain boundaries (GBs).

AB - Efficiently designing lightweight alloys with combined high corrosion resistance and mechanical properties remains an enduring topic in materials engineering. Due to the inadequate accuracy of conventional stress-strain machine learning (ML) models caused by corrosion factors, a novel reinforcement self-learning ML algorithm combined with calculated features (accuracy R2 >0.92) is developed. Based on the ML models, calculated work functions and mechanical moduli, a Computation Designed Corrosion-Resistant Al alloy is fabricated and verified. The performance (elongation reaches ∼30 %) is attributed to the H trapping Al-Sc-Cu phases (-1.44 eV H−1) and Cu-modified η/η' precipitates inside the grain boundaries (GBs).

KW - Al-Zn-Mg alloys

KW - First-principles calculation

KW - Machine learning

KW - Molecular dynamic simulation

KW - Precipitates

UR - http://www.scopus.com/inward/record.url?scp=85190507169&partnerID=8YFLogxK

U2 - 10.1016/j.corsci.2024.112062

DO - 10.1016/j.corsci.2024.112062

M3 - Article

AN - SCOPUS:85190507169

SN - 0010-938X

VL - 233

JO - Corrosion Science

JF - Corrosion Science

M1 - 112062

ER -

Artificial intelligence combined with high-throughput calculations to improve the corrosion resistance of AlMgZn alloy

Abstract

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this