Efficient machine-learning model for fast assessment of elastic properties of high-entropy alloys

We combined descriptor-based analytical models for stiffness-matrix and elastic-moduli with meanfield methods to accelerate assessment of technologically useful properties of high-entropy alloys, such as strength and ductility. Model training for elastic properties uses Sure-Independence Screening (SIS) and Sparsifying Operator (SO) method yielding an optimal analytical model, constructed with meaningful atomic features to predict target properties. Computationally inexpensive analytical descriptors were trained using a database of elastic properties determined from density functional theory for binary and ternary subsets of Nb-Mo-Ta-W-V refractory alloys. The optimal Elastic-SISSO models, extracted from an exponentially large feature space, give an extremely accurate prediction of target properties, similar to or better than other models, with some verified from existing experiments. We also show that electronegativity variance and elastic-moduli can directly predict trends in ductility and yield strength of refractory HEAs, and reveals promising alloy concentration regions.(c) 2022 Published by Elsevier Ltd on behalf of Acta Materialia Inc.

Automated summary

In this study, descriptor-based analytical models and meanfield methods were combined to accelerate the assessment of technologically useful properties of high-entropy alloys. The models accurately predicted target properties and revealed promising alloy concentration regions.