Journal
MATERIALS TODAY-PROCEEDINGS
Volume 38, Issue -, Pages 1535-1540Publisher
ELSEVIER
DOI: 10.1016/j.matpr.2020.08.145
Keywords
High entropy alloys; Machine learning; Yield strength
Categories
Funding
- Russian Science Foundation
Ask authors/readers for more resources
In this study, machine learning techniques were utilized to predict the yield strengths of high-entropy alloys at different temperatures. Experimental validation showed satisfactory accuracy of the prediction model, although the accuracy was lower for alloys outside the training dataset.
Experimental evaluations of mechanical properties and investigations microstructure are time-intensive, requiring weeks or months to produce and characterize a small number of candidate alloys. In this work, machine learning approaches were used for prediction yield strengths of high-entropy alloys Al-Cr-NbTi-V-Zr system at 20, 600 and 800 degrees C. Surrogate prediction model was built with support vector regression algorithm by a dataset including more 30 alloys Al-Cr-Nb-Ti-V-Zr system. Four model alloys were fabricated for testing the surrogate model by vacuum arc melting. After that model alloys were annealed in a quartz tube at 1200 degrees C 10 h. The microstructure of alloys after heat treatment were investigated with methods of scanning electron microscopy and X-ray structural analysis. Specimens of model alloys were compressed in the air at a nominal strain rate of 10 (4) s (1) at 20, 600 and 800 degrees C in a universal testing machine to determine the yield strength. The model showed the satisfactory accuracy prediction of yield strengths as single-phase as multi-phase alloys at all test temperatures. In connection with the small size of training dataset accuracy prediction of yield strengths for alloys outside composition space of training dataset is lower than inside. (C) 2020 Elsevier Ltd. All rights reserved.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available