Current application status of multi-scale simulation and machine learning in research on high-entropy alloys

Abstract

High-entropy alloys (HEAs) have garnered significant attention across various fields owing to their unique design incorporating multi-principal elements and remarkable comprehensive performance. Nevertheless, the enormous composition design space of HEAs makes conventional alloy design methods appear to be costly and inefficient. Recently, computer simulation technologies such as multi-scale simulation and machine learning have emerged as an efficient way to explore the composition design, structure, and performance simulation of HEAs.This review introduces the commonly used multi-scale simulation methods such as first-principles calculation, molecular dynamics simulation, Monte Carlo simulation, CALPHAD, finite element simulation, and machine learning. These methods not only simulate the microstructure and deformation behavior of HEAs but also predict crucial material properties like mechanical and physicochemical properties, thereby facilitating the design of HEAs. The current state-of-the-art advancements in multi-scale simulation and machine learning techniques for studying HEAs are summarized, encompassing their practical applications and potential limitations. The utilization of machine learning and multi-scale computation in materials science, as well as the future prospects are ultimately proposed.

Keywords

• Multi-scale simulation
• Machine learning
• High-entropy alloys
• Simulation and prediction
• Alloy design