|About this Abstract
||3rd World Congress on High Entropy Alloys (HEA 2023)
||Unleashing the Power of Machine Learning for High-Entropy Alloy Discovery: Phase Prediction
||Sima Alidokht, Ehsan Gerashi, Armin Hatefi
|On-Site Speaker (Planned)
High-entropy alloys (HEAs) have gained significant attention in materials science for their remarkable mechanical properties and extensive compositional versatility. Due to their high-dimensional chemical complexity, understanding their physical mechanisms and designing new HEAs is challenging. Predicting HEA phases can provide valuable insights, including mechanical properties anticipations. The conventional trial-and-error approach for discovering new HEAs is time-consuming and costly. To address the issue, we employ the power of machine learning methods to predict the phase of HEAs, reducing the effort required for HEA design. In this research, we propose various statistical and machine learning and artificial neural networks to model the HEAs phase responses, estimate the model parameters and explain the relationship between the HEAs phase features. Through extensive numerical experiments, we investigate the effects of design parameters in identifying various phases and evaluate the performance of the proposed models in estimating and predicting the HEA phases.
||Planned: Metallurgical and Materials Transactions