机器学习辅助设计具有增强硬度的crtativ基耐火高熵合金

IF 4.6 2区材料科学 Q2 MATERIALS SCIENCE, MULTIDISCIPLINARY

International Journal of Refractory Metals & Hard Materials Pub Date : 2025-08-22 DOI:10.1016/j.ijrmhm.2025.107400

Liang Li, Shu Wang, Chen Su, Shengfeng Guo

{"title":"机器学习辅助设计具有增强硬度的crtativ基耐火高熵合金","authors":"Liang Li, Shu Wang, Chen Su, Shengfeng Guo","doi":"10.1016/j.ijrmhm.2025.107400","DOIUrl":null,"url":null,"abstract":"<div><div>Refractory high entropy alloys (RHEAs) with enhanced hardness exhibit superior performance. However, traditional trial-and-error methods could be time-consuming and inefficient facing the enormous compositional space. In this paper, machine learning (ML) was utilized to design RHEAs with enhanced hardness. Several candidate alloys were selected by high-throughput screening and verified. The hardness of Cr<sub>45</sub>Ta<sub>21</sub>Ti<sub>20</sub>V<sub>14</sub> is up to 1074 HV, which is 33.4 % higher than the maximum value (805 HV) in the database. Moreover, the Shapley additive explanation (SHAP) was introduced to further comprehend the model interpretability.</div></div>","PeriodicalId":14216,"journal":{"name":"International Journal of Refractory Metals & Hard Materials","volume":"133 ","pages":"Article 107400"},"PeriodicalIF":4.6000,"publicationDate":"2025-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Machine learning assisted design of CrTaTiV-based refractory high entropy alloys with enhanced hardness\",\"authors\":\"Liang Li, Shu Wang, Chen Su, Shengfeng Guo\",\"doi\":\"10.1016/j.ijrmhm.2025.107400\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<div><div>Refractory high entropy alloys (RHEAs) with enhanced hardness exhibit superior performance. However, traditional trial-and-error methods could be time-consuming and inefficient facing the enormous compositional space. In this paper, machine learning (ML) was utilized to design RHEAs with enhanced hardness. Several candidate alloys were selected by high-throughput screening and verified. The hardness of Cr<sub>45</sub>Ta<sub>21</sub>Ti<sub>20</sub>V<sub>14</sub> is up to 1074 HV, which is 33.4 % higher than the maximum value (805 HV) in the database. Moreover, the Shapley additive explanation (SHAP) was introduced to further comprehend the model interpretability.</div></div>\",\"PeriodicalId\":14216,\"journal\":{\"name\":\"International Journal of Refractory Metals & Hard Materials\",\"volume\":\"133 \",\"pages\":\"Article 107400\"},\"PeriodicalIF\":4.6000,\"publicationDate\":\"2025-08-22\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"International Journal of Refractory Metals & Hard Materials\",\"FirstCategoryId\":\"88\",\"ListUrlMain\":\"https://www.sciencedirect.com/science/article/pii/S0263436825003658\",\"RegionNum\":2,\"RegionCategory\":\"材料科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q2\",\"JCRName\":\"MATERIALS SCIENCE, MULTIDISCIPLINARY\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"International Journal of Refractory Metals & Hard Materials","FirstCategoryId":"88","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0263436825003658","RegionNum":2,"RegionCategory":"材料科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"MATERIALS SCIENCE, MULTIDISCIPLINARY","Score":null,"Total":0}

引用次数: 0

摘要

硬度增强的难熔高熵合金（RHEAs）表现出优异的性能。然而，面对巨大的构图空间，传统的试错方法既费时又低效。本文利用机器学习（ML）来设计具有增强硬度的RHEAs。通过高通量筛选筛选了几种候选合金，并进行了验证。Cr45Ta21Ti20V14的硬度高达1074 HV，比数据库中的最大值805 HV高出33.4%。此外，引入Shapley加性解释（SHAP）进一步理解了模型的可解释性。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Machine learning assisted design of CrTaTiV-based refractory high entropy alloys with enhanced hardness

Refractory high entropy alloys (RHEAs) with enhanced hardness exhibit superior performance. However, traditional trial-and-error methods could be time-consuming and inefficient facing the enormous compositional space. In this paper, machine learning (ML) was utilized to design RHEAs with enhanced hardness. Several candidate alloys were selected by high-throughput screening and verified. The hardness of Cr₄₅Ta₂₁Ti₂₀V₁₄ is up to 1074 HV, which is 33.4 % higher than the maximum value (805 HV) in the database. Moreover, the Shapley additive explanation (SHAP) was introduced to further comprehend the model interpretability.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

$International Journal of Refractory Metals & Hard Materials$

International Journal of Refractory Metals & Hard Materials 工程技术-材料科学：综合

CiteScore

7.00

自引率

13.90%

发文量

236

审稿时长

35 days

期刊介绍： The International Journal of Refractory Metals and Hard Materials (IJRMHM) publishes original research articles concerned with all aspects of refractory metals and hard materials. Refractory metals are defined as metals with melting points higher than 1800 °C. These are tungsten, molybdenum, chromium, tantalum, niobium, hafnium, and rhenium, as well as many compounds and alloys based thereupon. Hard materials that are included in the scope of this journal are defined as materials with hardness values higher than 1000 kg/mm2, primarily intended for applications as manufacturing tools or wear resistant components in mechanical systems. Thus they encompass carbides, nitrides and borides of metals, and related compounds. A special focus of this journal is put on the family of hardmetals, which is also known as cemented tungsten carbide, and cermets which are based on titanium carbide and carbonitrides with or without a metal binder. Ceramics and superhard materials including diamond and cubic boron nitride may also be accepted provided the subject material is presented as hard materials as defined above.