Review on applications of artificial neural networks to develop high entropy alloys: A state-of-the-art technique

Abstract

Compared to conventional alloys, multicomponent high-entropy alloys (HEAs) have received considerable attention in recent years owing to their exceptional phase stability and mechanical properties. A detailed understanding of the interface between materials research and artificial intelligence has become critical for the perspective of developing advanced HEAs with desired properties. As the mechanical performance of HEAs is related to the phase composition and microstructure, the prediction of those characteristics becomes of immense interest to avoid complex experimental steps and reduce the time and manufacturing costs. In this context, machine learning-assisted artificial neural network (ANN) modeling is a computer-based method for developing novel materials by predicting potential alloying elements to tune the desired phase and material performance. The present review focuses on the application of ANN modeling in the prediction of the phase formation, microstructures, and mechanical properties of HEAs.