+高级检索
首页 > 过刊浏览>2023年第52卷第4期 >1192-1200. DOI:10.12442/j.issn.1002-185X.20220750
PDF HTML阅读 XML下载 导出引用 引用提醒
基于可解释机器学习模型的难熔高熵合金相预测
作者:
作者单位:

国防科技大学 空天科学学院 材料科学与工程系，湖南 长沙 410073

基金项目:

National Natural Science Foundation of China (Grant Nos. U20A20231 and 11972372) and College of Aerospace Science and Engineering Youth Talent Fund, National University of Defense Technology (Grant No. KY0505072209).


Interpretable Machine Learning Model-Based Phase Prediction for Refractory High-Entropy Alloys
Author:
Affiliation:

Department of Materials Science and Engineering, College of Aerospace Science and Engineering, National University of Defense Technology, Changsha 410073, China

Fund Project:

National Natural Science Foundation of China (U20A20231, 11972372); College of Aerospace Science and Engineering Youth Talent Fund, National University of Defense Technology (KY0505072209)

  • 摘要
    • | |
  • 访问统计
    • |
  • 参考文献 [36]
    • |
  • 相似文献 [20]
    • | | |
  • 文章评论
    摘要:

    采用k近邻 （KNN）、支持向量机（SVM）、决策树（DT）、随机森林（RF）和人工神经网络（ANN）5种机器学习（ML）方法对RHEAs中固溶体（SS）、混合固溶体和金属间化合物（SS+IM）进行了分类和预测。选择了5个输入相预测参数作为特征以及139组RHEAs数据以训练ML模型。结果表明，ANN模型的预测准确率最高，达到90.72%。9组新的四元和(TiVTa)xCr1–x体系RHEAs的实验结果显示，RF和ANN的预测精度更高，精准预测了11个SS和3个SS+IM合金的相组成。采用了SHAP（SHapley Additive exPlanations）模型来解释精度最高的 ANN 模型，并研究每个特征对相形成的贡献。5个特征的重要性顺序是混合焓（ΔHmix）、原子尺寸差（δ）、价电子浓度（VEC）、混合熵（ΔSmix）和电负性差（Δχ），其中ΔHmix的平均SHAP值大约是Δχ的5倍，是ΔSmix的4倍。较大的ΔHmix、较小的δ和VEC可能有助于RHEA中固溶体的形成。

    Abstract:

    Five machine learning (ML) approaches, i.e. K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Decision Tree (DT), Random Forest (RF) and Artificial Neural Network (ANN) were used to classify and to predict the combination of phases, i.e. solid solutions (SS) and mixed solid solution and intermetallic (SS+IM) in refractory high-entropy alloys (RHEAs). Five input characteristic phase predicting parameters and 139 RHEAs were selected to train these models. Results show that ANN model has the highest accuracy of 90.72%. Experimental results of 9 quaternary and (TiVTa)xCr1–x RHEAs verify the accuracy of prediction and indicate that RF and ANN can predict more accurately, successfully predicting 11 SS and 3 SS+IM. SHAP (SHapley Additive exPlanations) model was used to interpret the ANN model which exhibits the highest accuracy and to investigate the contribution of each feature to phase formation. The order of importance of five features is enthalpy of mixing (ΔHmix), atomic size difference (δ), valence electron concentration (VEC), entropy of mixing (ΔSmix), and electronegativity difference (Δχ), where the mean SHAP value of ΔHmix is approximately 5 times higher than that of ?χ and 4 times higher than that of ΔSmix. Less negative ΔHmix, smaller δ and VEC may contribute to the formation of SS in RHEAs.

    参考文献
    [1] Cantor B, Chang I, Knight P et al. Materials Science and Engineering A[J], 2004, 375?377: 213
    [2] Yeh J W, Chen S K, Lin S J et al. Advanced Engineering Materials[J], 2004, 6(5): 299
    [3] Senkov O N, Miracle D B. Acta Materialia[J], 2017, 122: 448
    [4] Lei Z F, Liu X J, Wu Y et al. Nature[J], 2018, 563(7732): 546
    [5] Senkov O N, Miracle D B, Chaput K J et al. Journal of Materials Research[J], 2018, 33(19): 3092
    [6] Rw A, Yu T A, Zlb C et al. Materials & Design[J], 2021, 213: 110 356
    [7] Wei S L, Kim S J, Kang J et al. Nature Materials[J], 2020, 19(11): 1175
    [8] An Z B, Mao S C, Yang Tao et al. Materials Horizons[J], 2021, 8(3): 948
    [9] Zhang Z R, Tang Y. Materials & Design[J], 2017, 133: 435
    [10] Luo J, Sun W, Duan R et al. Science & Technology[J], 2022, 110: 43
    [11] Atwani O El, Li N, Li M et al. Science Advances[J], 2019, 5(3): 2002
    [12] Zhang Z, Armstrong D, Grant P S. Progress in Materials Science[J], 2021, 123: 100 807
    [13] Chen S Y, Tong Y, Tseng K K et al. Scripta Materialia[J], 2019, 158: 50
    [14] Andreoli A F, Mendes R G, Witusiewicz V T et al. Acta Materialia[J], 2021, 221: 117 416
    [15] Senkov O N, Jensen J K, Pilchak A L et al. Materials & De- sign[J], 2018, 139: 498
    [16] Miracle D B, Senkov O N. Acta Materialia[J], 2017, 122: 448
    [17] Yang X, Zhang Y. Materials Chemistry and Physics[J], 2012, 132(2?3): 233
    [18] Wang J, Tang Y, Li S et al. Intermetallics[J], 2022, 142: 107 436
    [19] Guo S. Materials Science and Technology[J], 2015, 31(10): 1223
    [20] Zeng Y, Man M, Bai K et al. Materials & Design[J], 2021, 202(8): 109 532
    [21] Ju S P, Lee I J, Chen H Y. Journal of Alloys and Compounds[J], 2021, 858: 157 681
    [22] Rabin D, Meshulam A, Fuks D et al. Journal of Alloys and Compounds[J], 2021, 874: 159 940
    [23] Agarwal A, Rao A K P. JOM[J], 2019, 71(10): 3424
    [24] Zhou Z, Zhou Y, He Q et al. npj Computational Materials[J], 2019, 5: 128
    [25] Islam N, Huang Zhuang W, H L. Computational Materials Science[J], 2018, 150: 230
    [26] Krishna Y V, Jaiswal U K, Rahul M R. Scripta Materialia[J], 2021, 197: 113 804
    [27] Dai D, Xu T, Wei X et al. Computational Materials Science[J], 2020, 175: 109 618
    [28] Risal S, Zhu W, Guillen P et al. Computational Materials Sci-ence[J], 2021, 192: 110 389
    [29] Qi J, Cheung A M, Poon S J. Scientific Reports[J], 2019, 9: 15 001
    [30] Lundberg S, Lee S I. Conference and Workshop on Neural Information Processing Systems[C]. California: MIT Press, 2017: 4768
    [31] Senkov O N, Miracle D B, Chaput K J et al. Journal of Materials Research[J], 2018, 33(19): 3092
    [32] Pedregosa F, Varoquaux G, Gramfort A et al. Journal of Machine Learning Research[J], 2011, 12: 2825
    [33] Takahashi K, Takahashi L. Journal of Physical Chemistry Let-ters[J], 2019, 10(2): 283
    [34] Cutler D R, Edwards T C, Beard K H et al. Ecology[J], 2007, 88(11): 2783
    [35] Huang W, Martin P, Zhuang H L. Acta Materialia[J], 2019, 169: 225
    [36] Li Y, Tang Y, Ye Y et al. Rare Metal Materials and Engineer- ing[J], 2021, 50(5): 1635
    引证文献
    网友评论
    网友评论
    分享到微博
    发 布
引用本文

赵凤媛,叶益聪,张周然,李亚豪,王洁,唐宇,李顺,白书欣.基于可解释机器学习模型的难熔高熵合金相预测[J].稀有金属材料与工程,2023,52(4):1192~1200.[Zhao Fengyuan, Ye Yicong, Zhang Zhouran, Li Yahao, Wang Jie, Tang Yu, Li Shun, Bai Shuxin. Interpretable Machine Learning Model-Based Phase Prediction for Refractory High-Entropy Alloys[J]. Rare Metal Materials and Engineering,2023,52(4):1192~1200.]
DOI：10.12442/j. issn.1002-185X.20220750

复制
文章指标
  • 点击次数:711
  • 下载次数: 1319
  • HTML阅读次数: 275
  • 引用次数: 0
历史
  • 收稿日期:2022-09-20
  • 最后修改日期:2023-03-23
  • 录用日期:2022-12-23
  • 在线发布日期: 2023-04-28
  • 出版日期: 2023-04-25

总访问量 36088671

地址：西安市未央区未央路96号 邮政编码：710016 联系电话：029－86231117

E-mail：rmme@c-nin.com; rmme0626@aliyun.com

版权所有:稀有金属材料与工程 ® 2025 版权所有 技术支持:北京勤云科技发展有限公司 ICP:陕ICP备05006818号-3