Application of Machine Learning to Mechanical Properties of Copper Graphene Composites

TL;DR

This study applies machine learning models to predict the mechanical properties of copper-graphene composites, providing insights into key influencing factors and demonstrating satisfactory predictive accuracy.

Contribution

It introduces the use of multiple ML models for predicting properties of Cu/Gr composites and identifies key factors affecting their strength and conductivity.

Findings

ML models predict composite properties with satisfactory accuracy

Feature analysis reveals key factors influencing mechanical properties

Four different ML models were evaluated for this task

Abstract

While copper-graphene (Cu/Gr) composites have been promising materials due to their theoretically high strength and conductivity, their design has been hampered by the large number of variables affecting their properties. We applied four different Machine Learning (ML) models to manually collected datasets compiling the yield strength and ultimate tensile strength of graphene-reinforced copper composites processed with powder metallurgy techniques. Our results indicate that ML models can predict the mechanical properties of Cu/Gr composites with satisfactory accuracy. Feature analysis provided new insights into the most important factors that affect these properties.