Jiahao Qian, Yang Li, Jialiang Hou, Shaojie Wu, Yun Zou
{"title":"通过基于贝叶斯优化的深度学习加速开发具有高硬度的铁-铜-镍-铬体系 HEAs","authors":"Jiahao Qian, Yang Li, Jialiang Hou, Shaojie Wu, Yun Zou","doi":"10.1557/s43578-024-01370-0","DOIUrl":null,"url":null,"abstract":"<p>Machine learning (ML) is progressively supplanting conventional trial-and-error approaches for designing alloys with desirable properties. In this study, four ML regression models were utilized to identify high-entropy alloys (HEAs) with high hardness within the Fe–Co–Ni–Cr system. The Bayesian optimized deep learning (BO-DL) method yielded the highest prediction accuracy (<i>R</i><sup><i>2</i></sup> = 0.93). Notably, the BO-DL method is no longer limited to a single HEA system and now can target different alloy systems composed of more elements with reliable prediction results. Furthermore, a genetic algorithm was utilized to search for HEAs with high hardness. The accuracy and reliability of the predictions were experimentally verified. As-cast Fe<sub>5</sub>Co<sub>20</sub>Ni<sub>10</sub>Cr<sub>30</sub>Al<sub>5</sub>Ti<sub>30</sub> HEA exhibited a remarkable hardness of 890 HV, which is one of the highest for alloys in the Fe–Co–Ni–Cr system. The methodologies and framework proposed in this study can serve as a blueprint for facilitating the design of HEAs.</p><h3 data-test=\"abstract-sub-heading\">Graphical Abstract</h3>\n","PeriodicalId":16306,"journal":{"name":"Journal of Materials Research","volume":null,"pages":null},"PeriodicalIF":2.7000,"publicationDate":"2024-06-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Accelerating the development of Fe–Co–Ni–Cr system HEAs with high hardness by deep learning based on Bayesian optimization\",\"authors\":\"Jiahao Qian, Yang Li, Jialiang Hou, Shaojie Wu, Yun Zou\",\"doi\":\"10.1557/s43578-024-01370-0\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>Machine learning (ML) is progressively supplanting conventional trial-and-error approaches for designing alloys with desirable properties. In this study, four ML regression models were utilized to identify high-entropy alloys (HEAs) with high hardness within the Fe–Co–Ni–Cr system. The Bayesian optimized deep learning (BO-DL) method yielded the highest prediction accuracy (<i>R</i><sup><i>2</i></sup> = 0.93). Notably, the BO-DL method is no longer limited to a single HEA system and now can target different alloy systems composed of more elements with reliable prediction results. Furthermore, a genetic algorithm was utilized to search for HEAs with high hardness. The accuracy and reliability of the predictions were experimentally verified. As-cast Fe<sub>5</sub>Co<sub>20</sub>Ni<sub>10</sub>Cr<sub>30</sub>Al<sub>5</sub>Ti<sub>30</sub> HEA exhibited a remarkable hardness of 890 HV, which is one of the highest for alloys in the Fe–Co–Ni–Cr system. The methodologies and framework proposed in this study can serve as a blueprint for facilitating the design of HEAs.</p><h3 data-test=\\\"abstract-sub-heading\\\">Graphical Abstract</h3>\\n\",\"PeriodicalId\":16306,\"journal\":{\"name\":\"Journal of Materials Research\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":2.7000,\"publicationDate\":\"2024-06-18\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Materials Research\",\"FirstCategoryId\":\"88\",\"ListUrlMain\":\"https://doi.org/10.1557/s43578-024-01370-0\",\"RegionNum\":4,\"RegionCategory\":\"材料科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"MATERIALS SCIENCE, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Materials Research","FirstCategoryId":"88","ListUrlMain":"https://doi.org/10.1557/s43578-024-01370-0","RegionNum":4,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"MATERIALS SCIENCE, MULTIDISCIPLINARY","Score":null,"Total":0}
引用次数: 0
摘要
在设计具有理想特性的合金时,机器学习(ML)正逐步取代传统的试错法。本研究利用四种 ML 回归模型来识别铁-铜-镍-铬体系中具有高硬度的高熵合金 (HEA)。贝叶斯优化深度学习(BO-DL)方法的预测精度最高(R2 = 0.93)。值得注意的是,BO-DL 方法不再局限于单一的 HEA 系统,现在可以针对由更多元素组成的不同合金系统得出可靠的预测结果。此外,还利用遗传算法来寻找高硬度的 HEA。实验验证了预测的准确性和可靠性。铸件 Fe5Co20Ni10Cr30Al5Ti30 HEA 的硬度高达 890 HV,是铁-钴-镍-铬体系中硬度最高的合金之一。本研究提出的方法和框架可作为促进 HEA 设计的蓝图。
Accelerating the development of Fe–Co–Ni–Cr system HEAs with high hardness by deep learning based on Bayesian optimization
Machine learning (ML) is progressively supplanting conventional trial-and-error approaches for designing alloys with desirable properties. In this study, four ML regression models were utilized to identify high-entropy alloys (HEAs) with high hardness within the Fe–Co–Ni–Cr system. The Bayesian optimized deep learning (BO-DL) method yielded the highest prediction accuracy (R2 = 0.93). Notably, the BO-DL method is no longer limited to a single HEA system and now can target different alloy systems composed of more elements with reliable prediction results. Furthermore, a genetic algorithm was utilized to search for HEAs with high hardness. The accuracy and reliability of the predictions were experimentally verified. As-cast Fe5Co20Ni10Cr30Al5Ti30 HEA exhibited a remarkable hardness of 890 HV, which is one of the highest for alloys in the Fe–Co–Ni–Cr system. The methodologies and framework proposed in this study can serve as a blueprint for facilitating the design of HEAs.
期刊介绍:
Journal of Materials Research (JMR) publishes the latest advances about the creation of new materials and materials with novel functionalities, fundamental understanding of processes that control the response of materials, and development of materials with significant performance improvements relative to state of the art materials. JMR welcomes papers that highlight novel processing techniques, the application and development of new analytical tools, and interpretation of fundamental materials science to achieve enhanced materials properties and uses. Materials research papers in the following topical areas are welcome.
• Novel materials discovery
• Electronic, photonic and magnetic materials
• Energy Conversion and storage materials
• New thermal and structural materials
• Soft materials
• Biomaterials and related topics
• Nanoscale science and technology
• Advances in materials characterization methods and techniques
• Computational materials science, modeling and theory