Interpretable machine learning workflow for evaluation of the transformation temperatures of TiZrHfNiCoCu high entropy shape memory alloys

• DOI: 10.1016/j.matdes.2022.111513
• Publication Date: 2022-12-17T07:02:07Z
• Authors: Shiyu He; Yanming Wang; Zhengyang Zhang; Fei Xiao; Shungui Zuo; Ying Zhou; Xiaorong Cai; Xuejun Jin
• Keywords: Electronegativity; High-Entropy Alloys; Univariate

Machine learning approaches (ML) based on data-driven models are conducive to accelerating the assessments of martensitic transformation peak temperature (Tp) TiZrHfNiCoCu high entropy shape memory alloys (HESMAs) over a huge composition space. In this work, an interpretable machine workflow was established through dataset construction, feature selection, modeling and validation, model interpretation. We identified set key combinations closely related Tp, by exploiting Pearson correlation univariate forward elimination. The ML then used estimate Tp three newly synthesized alloys, with their prediction relative errors less than 3 % in comparison experimental measurements. behaviors our were interpreted Shapley Additive exPlainations (SHAP) approach, demonstrating crucial role CV22 (Allred Rochow electronegativity) Tp. addition, combination designed interpretation strategy further investigate effects alloying elements which showed that HESMAs 9 ≦ Co (mol%) ≤ 10 15 Cu 17 have pronounced positive