Advances in machine learning methods in copper alloys: a review

Abstract

Context

Advanced copper and copper alloys, as significant engineering structural materials, have recently been extensively used in energy, electron, transportation, and aviation domains. Higher requirements urge the emergence of high-performance copper alloys. However, the traditional trial-and-error experimental observations and computational simulation research used to design and develop novel materials are time-consuming and costly. With the accumulation of material research and rapid development of computational ability, the thorough application of material genome engineering has sped up the development of novel materials and facilitates the process of systematic engineering application.

Methods

This review summarizes the benefits of data-driven machine learning techniques and the state of the art of machine learning research in the area of copper alloys. It also displays the widely used computational simulation approaches (e.g., the first-principles calculation, molecular dynamics simulation, phase-field simulations, and finite element analysis) and their combined applications in material design and property prediction. Finally, the limitations of machine learning research methods are outlined, and future development directions are proposed.