Scholarlink

<p>High-entropy alloys (HEAs), especially refractory HEAs (RHEAs), are vital for high-temperature applications due to their superior mechanical properties. This study predicts the hardness of MoNbTaW, MoNbTiW, and MoNbTaTiW RHEAs using Density Functional Theory (DFT), machine learning (ML), and experiments. Atomic structures were generated <i>via</i> ATAT, and hardness was computed using DFT. Eleven ML models, including Random Forest and XGBoost, were trained on five empirical descriptors. Experimental validation involved XRD, SEM, and hardness testing. Random Forest and XGBoost performed best (<i>R</i><sup>2</sup> = 0.89–0.90), identifying valence electron concentration and mixing entropy as key predictors.</p> Graphical abstract <p></p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Integrating machine learning and DFT for hardness prediction in high-entropy alloys

  • 期刊论文

  • 发表日期：

    2025年5月22日

查看全文
  • Shanker Kumar,
  • Vikas Jindal

摘要

High-entropy alloys (HEAs), especially refractory HEAs (RHEAs), are vital for high-temperature applications due to their superior mechanical properties. This study predicts the hardness of MoNbTaW, MoNbTiW, and MoNbTaTiW RHEAs using Density Functional Theory (DFT), machine learning (ML), and experiments. Atomic structures were generated via ATAT, and hardness was computed using DFT. Eleven ML models, including Random Forest and XGBoost, were trained on five empirical descriptors. Experimental validation involved XRD, SEM, and hardness testing. Random Forest and XGBoost performed best (R2 = 0.89–0.90), identifying valence electron concentration and mixing entropy as key predictors.

Graphical abstract