journal article·2023-11-05Open access

Predicting the Hardness of Al-Sc-X Alloys with Machine Learning Models, Explainable Artificial Intelligence Analysis and Inverse Design

Jiwon Park, Su-Hyeon Kim, Jisu Kim, Byung-joo Kim, Hyun-seok Cheon, Chang-Seok Oh

Abstract

In this study, the Vickers hardness of precipitation-strengthened Al-Sc-X (X = Zr, Si, and Fe) alloys were predicted using machine learning models, depending on the alloys’ compositions, solid-solution treatment and aging conditions. The data used for machine learning were collected from the literature. Among the models, tree-based ensemble models such as extreme gradient boosting and random forest performed well. Then the feature impact on the model output was analyzed with SHarpely Additive eXplanation (SHAP). Based on the SHAP analysis and prior domain knowledge, the process conditions were restricted to narrow down the inverse design search space. Candidate alloys suggested by the optimization using a genetic algorithm showed improved hardness values. The hardness prediction model and the inverse designsuggested candidates were then experimentally validated. The accuracy of the hardness prediction model was 0.994, when the predicted hardness was 85.4 Hv, and the experimentally measured hardness was 84.9 Hv. A specimen whose composition was close to the inverse-designed alloy was cast and heat treated according to the suggested conditions. The inverse design showed an accuracy of 0.965. Exploring the entire combination of possible feature space requires vast effort and time. An efficient search for materials with improved properties can be achieved using an appropriate configuration of well-performing machine learning models and explainable AI techniques guided by domain knowledge.
Source: crossref · Quality: baseline

Research context

OpenAlex topics: Hydrogen embrittlement and corrosion behaviors in metals · Microstructure and Mechanical Properties of Steels · Advanced Materials Characterization Techniques

SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.

References (39)

Show source-provided references
1. 10.1016/j.msea.2010.02.02710.1016/j.msea.2010.02.027 ↗
2. 10.1016/j.actamat.2010.10.01710.1016/j.actamat.2010.10.017 ↗
3. 10.1016/j.actamat.2010.11.03710.1016/j.actamat.2010.11.037 ↗
4. 10.1134/s207511331105017010.1134/s2075113311050170 ↗
5. 10.1016/j.actamat.2012.05.03610.1016/j.actamat.2012.05.036 ↗
6. 10.1016/j.matdes.2012.06.01910.1016/j.matdes.2012.06.019 ↗
7. Tuan · 2014 · 77
8. Zhang · 2014
9. Luca · 2016 · 35
10. 10.1016/j.msea.2016.09.06510.1016/j.msea.2016.09.065 ↗
11. 10.1016/j.jallcom.2017.02.04310.1016/j.jallcom.2017.02.043 ↗
12. 10.1016/j.msea.2016.10.06710.1016/j.msea.2016.10.067 ↗
13. 10.1590/1980-5373-mr-2017-048110.1590/1980-5373-mr-2017-0481 ↗
14. 10.1016/j.jallcom.2017.11.22410.1016/j.jallcom.2017.11.224 ↗
15. 10.1016/j.jmrt.2017.05.00610.1016/j.jmrt.2017.05.006 ↗
16. 10.1016/j.matchar.2019.03.04410.1016/j.matchar.2019.03.044 ↗
17. 10.1016/j.jallcom.2020.15582610.1016/j.jallcom.2020.155826 ↗
18. 10.1016/j.actamat.2007.11.01110.1016/j.actamat.2007.11.011 ↗
19. 10.1016/0022-3697(61)90054-310.1016/0022-3697(61)90054-3 ↗
20. Wagner · 1961 · 581
21. 10.1016/0956-716x(93)90191-t10.1016/0956-716x(93)90191-t ↗
22. 10.1007/s12540-022-01312-710.1007/s12540-022-01312-7 ↗
23. 10.1016/j.commatsci.2023.11224910.1016/j.commatsci.2023.112249 ↗
24. 10.1016/j.jallcom.2023.17035910.1016/j.jallcom.2023.170359 ↗
25. 10.1007/s12540-020-00809-310.1007/s12540-020-00809-3 ↗
26. 10.1039/d2ra05889h10.1039/d2ra05889h ↗
27. Lee · 2020 · 20443
28. Lee · 2021 · 11012
29. 10.3365/kjmm.2021.59.11.83810.3365/kjmm.2021.59.11.838 ↗
30. 10.3365/kjmm.2022.60.8.61910.3365/kjmm.2022.60.8.619 ↗
31. 10.1016/j.jmrt.2023.06.20710.1016/j.jmrt.2023.06.207 ↗
32. 10.1016/j.actamat.2022.11830210.1016/j.actamat.2022.118302 ↗
33. 10.1016/j.scriptamat.2022.11496510.1016/j.scriptamat.2022.114965 ↗
34. 2022
35. 2017
36. 2022
37. Ribeiro · 2016 · 1135
38. 2017
39. 10.1038/s42256-019-0138-910.1038/s42256-019-0138-9 ↗