基于机器学习的低合金钢大气腐蚀速率预测

匡惊鸥; 龙志林

doi:10.1007/s12613-023-2679-5

Volume 31 Issue 2

Feb. 2024

Turn off MathJax

图(11) / 表(12)

数据统计

计量

文章访问数: 932
HTML全文浏览量: 150
PDF下载量: 67
被引次数: 0

文章导航 > 矿物冶金与材料学报（英文） > 2024 > 31(2): 337-350

Jingou Kuangand Zhilin Long, Prediction model for corrosion rate of low-alloy steels under atmospheric conditions using machine learning algorithms, Int. J. Miner. Metall. Mater., 31(2024), No. 2, pp. 337-350. https://doi.org/10.1007/s12613-023-2679-5

Cite this article as:

Jingou Kuangand Zhilin Long, Prediction model for corrosion rate of low-alloy steels under atmospheric conditions using machine learning algorithms, Int. J. Miner. Metall. Mater., 31(2024), No. 2, pp. 337-350. https://doi.org/10.1007/s12613-023-2679-5

Citation:

PDF( 2263 KB)

引用本文 PDF XML SpringerLink

研究论文

基于机器学习的低合金钢大气腐蚀速率预测

湘潭大学土木工程学院, 湘潭 411101, 中国

通讯作者:
龙志林 E-mail: longzl@xtu.edu.cn

文章亮点

(1) 通过特征选择和属性转换，提高了腐蚀速率预测模型的精度

(2) 输入特征和目标属性之间的隐性关系可以使用SHAP方法来解释

(3) 特征选择方法可以有效地筛选出与腐蚀速率相关的关键特征

中文摘要

本文以LAS的材料特性、环境因素及暴露时间为输入，以腐蚀速率为输出，使用6种不同的机器学习(ML)模型来预测低合金钢(LAS)的大气腐蚀速率。通过超参数调整和特征筛选，极端梯度提升(XGBoost)模型表现最优的预测精度。随后使用属性转换方法，将材料性质特征转换为对应原子和物理特征，并使用递归特征剔除(RFE)方法和XGBoost特征筛选法，筛选出影响腐蚀速率的重要因素。使用属性转换特征所建立的ML模型表现出优异的预测性能和泛化能力。此外，还应用Shapley additive exPlanations(SHAP)方法分析了输入特征与腐蚀速率之间的关系。结果表明，属性转换模型能有效地帮助分析腐蚀行为，并显著提高腐蚀速率预测模型的泛化能力。

关键词:

Research Article

Prediction model for corrosion rate of low-alloy steels under atmospheric conditions using machine learning algorithms

Jingou Kuang and
Zhilin Long

+ Author Affiliations

Abstract

Abstract

This work constructed a machine learning (ML) model to predict the atmospheric corrosion rate of low-alloy steels (LAS). The material properties of LAS, environmental factors, and exposure time were used as the input, while the corrosion rate as the output. 6 different ML algorithms were used to construct the proposed model. Through optimization and filtering, the eXtreme gradient boosting (XGBoost) model exhibited good corrosion rate prediction accuracy. The features of material properties were then transformed into atomic and physical features using the proposed property transformation approach, and the dominant descriptors that affected the corrosion rate were filtered using the recursive feature elimination (RFE) as well as XGBoost methods. The established ML models exhibited better prediction performance and generalization ability via property transformation descriptors. In addition, the SHapley additive exPlanations (SHAP) method was applied to analyze the relationship between the descriptors and corrosion rate. The results showed that the property transformation model could effectively help with analyzing the corrosion behavior, thereby significantly improving the generalization ability of corrosion rate prediction models.
- machine learning;
- low-alloy steel;
- atmospheric corrosion prediction;
- corrosion rate;
- feature fusion

FullText(HTML)

Supplements(0)

References(38)

References

[1]	Z.F. Wang, J.R. Liu, L.X. Wu, R.D. Han, and Y.Q. Sun, Study of the corrosion behavior of weathering steels in atmospheric environments, Corros. Sci., 67(2013), p. 1. doi: 10.1016/j.corsci.2012.09.020
[2]	H. Cano, I. Díaz, D. de la Fuente, B. Chico, and M. Morcillo, Effect of Cu, Cr and Ni alloying elements on mechanical properties and atmospheric corrosion resistance of weathering steels in marine atmospheres of different aggressivities, Mater. Corros., 69(2018), No. 1, p. 8. doi: 10.1002/maco.201709656
[3]	W. Wu, L.L. Zhu, P.L. Chai, et al., Atmospheric corrosion behavior of Nb- and Sb-added weathering steels exposed to the South China Sea, Int. J. Miner. Metall. Mater., 29(2022), No. 11, p. 2041. doi: 10.1007/s12613-021-2383-2
[4]	M.H. Sun, X.J. Yang, C.W. Du, et al., Distinct beneficial effect of Sn on the corrosion resistance of Cr–Mo low alloy steel, J. Mater. Sci. Technol., 81(2021), p. 175. doi: 10.1016/j.jmst.2020.12.014
[5]	S. Jiang, J.P. Cao, Z.Y. Liu, X.X. Xu, J.W. Yang, and X.T. Li, Effect of Ni on the oxidation behavior of corrosion products that form on low alloy steel exposed to a thin electrolyte layer environment, Corros. Sci., 206(2022), art. No. 110471. doi: 10.1016/j.corsci.2022.110471
[6]	T. Kamimura, K. Kashima, K. Sugae, H. Miyuki, and T. Kudo, The role of chloride ion on the atmospheric corrosion of steel and corrosion resistance of Sn-bearing steel, Corros. Sci., 62(2012), p. 34. doi: 10.1016/j.corsci.2012.04.049
[7]	T.Q. Wu, M.C. Yan, J. Xu, Y.X. Liu, C. Sun, and W. Ke, Mechano-chemical effect of pipeline steel in microbiological corrosion, Corros. Sci., 108(2016), p. 160. doi: 10.1016/j.corsci.2016.03.011
[8]	C.G. Soares, Y. Garbatov, A. Zayed, and G. Wang, Influence of environmental factors on corrosion of ship structures in marine atmosphere, Corros. Sci., 51(2009), No. 9, p. 2014. doi: 10.1016/j.corsci.2009.05.028
[9]	H.Y. Tian, Z.Y. Cui, H. Ma, et al., Corrosion evolution and stress corrosion cracking behavior of a low carbon bainite steel in the marine environments: Effect of the marine zones, Corros. Sci., 206(2022), art. No. 110490. doi: 10.1016/j.corsci.2022.110490
[10]	A. Lazareva, J. Owen, S. Vargas, R. Barker, and A. Neville, Investigation of the evolution of an iron carbonate layer and its effect on localized corrosion of X65 carbon steel in CO₂ corrosion environments, Corros. Sci., 192(2021), art. No. 109849. doi: 10.1016/j.corsci.2021.109849
[11]	Y.M. Panchenko and A.I. Marshakov, Long-term prediction of metal corrosion losses in atmosphere using a power-linear function, Corros. Sci., 109(2016), p. 217. doi: 10.1016/j.corsci.2016.04.002
[12]	Y.K. Cai, Y. Zhao, X.B. Ma, K. Zhou, and Y. Chen, Influence of environmental factors on atmospheric corrosion in dynamic environment, Corros. Sci., 137(2018), p. 163. doi: 10.1016/j.corsci.2018.03.042
[13]	X.G. Sun, P. Lin, C. Man, et al., Prediction model for atmospheric corrosion of 7005-T4 aluminum alloy in industrial and marine environments, Int. J. Miner. Metall. Mater., 25(2018), No. 11, p. 1313. doi: 10.1007/s12613-018-1684-6
[14]	I. Díaz, H. Cano, B. Chico, D. de la Fuente, and M. Morcillo, Some clarifications regarding literature on atmospheric corrosion of weathering steels, Int. J. Corros., 2012(2012), art. No. 812192.
[15]	M. Morcillo, B. Chico, I. Díaz, H. Cano, and D. de la Fuente, Atmospheric corrosion data of weathering steels. A review, Corros. Sci., 77(2013), p. 6. doi: 10.1016/j.corsci.2013.08.021
[16]	B. Chico, D. De la Fuente, I. Díaz, J. Simancas, and M. Morcillo, Annual atmospheric corrosion of carbon steel worldwide. an integration of ISOCORRAG, ICP/UNECE and MICAT databases, Materials, 10(2017), No. 6, art. No. 601. doi: 10.3390/ma10060601
[17]	Y.J. Zhi, Z.H. Jin, L. Lu, et al., Improving atmospheric corrosion prediction through key environmental factor identification by random forest-based model, Corros. Sci., 178(2021), art. No. 109084. doi: 10.1016/j.corsci.2020.109084
[18]	Y.J. Zhi, D.M. Fu, D.W. Zhang, T. Yang, and X.G. Li, Prediction and knowledge mining of outdoor atmospheric corrosion rates of low alloy steels based on the random forests approach, Metals, 9(2019), No. 3, art. No. 383. doi: 10.3390/met9030383
[19]	L.C. Yan, Y.P. Diao, Z.Y. Lang, and K.W. Gao, Corrosion rate prediction and influencing factors evaluation of low-alloy steels in marine atmosphere using machine learning approach, Sci. Technol. Adv. Mater., 21(2020), No. 1, p. 359. doi: 10.1080/14686996.2020.1746196
[20]	Y.J. Zhi, T. Yang, and D.M. Fu, An improved deep forest model for forecast the outdoor atmospheric corrosion rate of low-alloy steels, J. Mater. Sci. Technol., 49(2020), p. 202. doi: 10.1016/j.jmst.2020.01.044
[21]	Y.J. Lv, J.W. Wang, J.L. Wang, et al., Steel corrosion prediction based on support vector machines, Chaos Solitons Fractals, 136(2020), art. No. 109807. doi: 10.1016/j.chaos.2020.109807
[22]	X. Wei, D.M. Fu, M.D. Chen, W. Wu, D.Q. Wu, and C. Liu, Data mining to effect of key alloying elements on corrosion resistance of low alloy steels in Sanya seawater environment alloying elements, J. Mater. Sci. Technol., 64(2021), p. 222. doi: 10.1016/j.jmst.2020.01.040
[23]	Z.B. Pei, D.W. Zhang, Y.J. Zhi, et al., Towards understanding and prediction of atmospheric corrosion of an Fe/Cu corrosion sensor via machine learning, Corros. Sci., 170(2020), art. No. 108697. doi: 10.1016/j.corsci.2020.108697
[24]	X.J. Yang, J.K. Yang, Y. Yang, et al., Data-mining and atmospheric corrosion resistance evaluation of Sn- and Sb-additional low alloy steel based on big data technology, Int. J. Miner. Metall. Mater., 29(2022), No. 4, p. 825. doi: 10.1007/s12613-022-2457-9
[25]	M. Kamrunnahar and M. Urquidi-Macdonald, Prediction of corrosion behavior using neural network as a data mining tool, Corros. Sci., 52(2010), No. 3, p. 669. doi: 10.1016/j.corsci.2009.10.024
[26]	J.C. Xie and L. Zhang, Machine learning and symbolic regression for adsorption of atmospheric molecules on low-dimensional TiO₂, Appl. Surf. Sci., 597(2022), art. No. 153728. doi: 10.1016/j.apsusc.2022.153728
[27]	B.Y. Ren, Z.L. Long, and R.J. Deng, A new criterion for predicting the glass-forming ability of alloys based on machine learning, Comput. Mater. Sci., 189(2021), art. No. 110259. doi: 10.1016/j.commatsci.2020.110259
[28]	Y.P. Diao, L.C. Yan, and K.W. Gao, Improvement of the machine learning-based corrosion rate prediction model through the optimization of input features, Mater. Des., 198(2021), art. No. 109326. doi: 10.1016/j.matdes.2020.109326
[29]	National Material Environment Corrosion Platform, National Material Corrosion and Protection Data Center, [2022-06-10]. https://www.corrdata.org.cn/
[30]	M. Shiri and D. Rezakhani, Estimated and stationary atmospheric corrosion rate of carbon steel, galvanized steel, copper and aluminum in Iran, Metall. Mater. Trans. A, 51(2020), No. 1, p. 342. doi: 10.1007/s11661-019-05509-1
[31]	Stratmann M. The atmospheric corrosion of iron—A discussion of the physico–chemical fundamentals of this omnipresent corrosion process, Ber. Bunsenges. Phys. Chem., 94(1990), No. 6, p. 626.
[32]	E. Bernardi, I. Vassura, S. Raffo, et al., Influence of inorganic anions from atmospheric depositions on weathering steel corrosion and metal release, Constr. Build. Mater., 236(2020), art. No. 117515. doi: 10.1016/j.conbuildmat.2019.117515
[33]	M. Morcillo, B. Chico, J. Alcántara, I. Díaz, J. Simancas, and D. de la Fuente, Atmospheric corrosion of mild steel in chloride-rich environments. Questions to be answered, Mater. Corros., 66(2015), No. 9, p. 882. doi: 10.1002/maco.201407940
[34]	P. Refait, O. Benali, M. Abdelmoula, and J.M.R. Génin, Formation of ‘ferric green rust’ and/or ferrihydrite by fast oxidation of iron (II–III) hydroxychloride green rust, Corros. Sci., 45(2003), No. 11, p. 2435. doi: 10.1016/S0010-938X(03)00073-8
[35]	Q.C. Zhang, J.S. Wu, J.J. Wang, W.L. Zheng, J.G. Chen, and A.B. Li, Corrosion behavior of weathering steel in marine atmosphere, Mater. Chem. Phys., 77(2003), No. 2, p. 603. doi: 10.1016/S0254-0584(02)00110-4
[36]	The University of Sheffield and WebElements Ltd, UK, The periodic table of the elements, [2022–06–27]. https://webelements.com/.
[37]	J.H. Chu, L.B. Tong, W. Wang, et al., Sequentially bridged biomimetic graphene-based coating via covalent bonding with an effective anti-corrosion/wear protection for Mg alloy, Colloids Surf. A, 610(2021), art. No. 125707. doi: 10.1016/j.colsurfa.2020.125707
[38]	C.Z. Wang, X.H. Wang, and X.Z. Qin, Research on atmospheric corrosion rule of carbon steel and low alloy steel in Chongqing and Wanning area, Equip. Environ. Eng., 3(2006), No. 2, p. 23.