Design, Performance Optimization and Application of High-Entropy Alloy Materials in Mechanical Engineering: A Review

Abstract

High-entropy alloys have multi-scale and complex microstructures, and their properties are highly tunable. They have great potential for development. However, the current development of high-entropy alloys is still dominated by trial and error, lacking effective guidance and low development efficiency. Machine learning is a data-based material design technology, which has been applied to the prediction of phase composition, prediction and optimization of mechanical properties, and auxiliary simulation calculations in the field of high-entropy alloys. However, the insufficiency of existing data, unbalanced distribution and the limitation of the model itself lead to great uncertainty in the composition optimization strategy based on machine learning. Based on this, this paper takes the machine learning method as the core, combines the composition design and the material design idea based on machine learning, and discusses its design idea in the high-entropy alloy system. And summarize their application research status in high entropy alloy composition screening, phase and structure calculation, and performance prediction. Finally, the current problems in this field are proposed, and solutions and future prospects are provided.