Lattice Misfit Predictions via the Gaussian Process Regression for Ni-Based Single Crystal Superalloys

• Published in: Metals and materials international Vol. 27; no. 2; pp. 235 - 253
• Main Authors: Zhang, Yun, Xu, Xiaojie
• Format: Journal Article
• Language: English
• Published: Seoul The Korean Institute of Metals and Materials 01.02.2021; Springer Nature B.V; 대한금속·재료학회
• Subjects: Characterization and Evaluation of Materials; Chemical composition; Chemistry and Materials Science; Cost analysis; Creep strength; Crystal lattices; Deformation mechanisms; Engineering Thermodynamics; Gaussian process; Heat and Mass Transfer; High temperature; Machines; Magnetic Materials; Magnetism; Manufacturing; Materials Science; Mechanical properties; Metallic Materials; Nickel; Nickel base alloys; Processes; Regression analysis; Regression models; Single crystals; Solid Mechanics; Superalloys; 재료공학; Ni-based single crystal superalloys; Lattice constant; Lattice mismatch; Machine learning
• ISSN: 1598-9623; 2005-4149
• DOI: 10.1007/s12540-020-00883-7

Ni-based single crystal superalloys exhibit superb mechanical strength, particularly, creep resistance at elevated temperature. The unique microstructure, which is consisted of γ and γ ′ phases, is a major factor that determines the mechanical behavior of these alloys. The lattice misfit between the two phases is of particular interest in understanding and predicting the deformation mechanism. The measurement of the lattice misfit by advanced analytical instruments is costly and difficult. In current study, we develop the Gaussian process regression model to predict lattice misfits for Ni-based single crystal superalloys based on chemical composition, temperature, and two morphological indicators. The model is highly stable and accurate and promising as a fast, robust, and low-cost tool for lattice misfit estimations. Graphic Abstract