|About this Abstract
||MS&T23: Materials Science & Technology
||Computational Discovery, Understanding, and Design of Multi-principal Element Materials
||Hybrid Machine Learning Approach for Designing Refractory High Entropy Alloys
||Debasis Sengupta, Stephen Giles, Hugh Shortt, Peter Liaw
|On-Site Speaker (Planned)
Development of process-structure-property relationships in materials science is an important and challenging frontier which promises improved materials and reduced time and cost in production. Refractory high entropy alloys (RHEAs) are a class of materials that are capable of excellent high-temperature properties. However, due to their multi-component nature, RHEAs have a vast composition space which presents challenges for traditional experimental exploration. In this work, we have used a number of machine learning method to predict room and high temperature yield strengths of RHEAs. The predicted results are also validated against experimental synthesis and characterization. However, our validation showed that no one method is consistently superior to the others. This work develops a novel graph-based hybrid method to intelligently combine the predictions of a number of machine learning methods. We demonstrated that the predictions of the hybrid method are superior to the all methods used in this work.