Predictive analytics of wear performance in high entropy alloy coatings through machine learning

Abstract High-entropy alloys (HEAs) are increasingly renowned for their distinct microstructural compositions and exceptional properties. These HEAs are employed for surface modification as coatings exhibit phenomenal mechanical characteristics including wear and corrosion resistance which are extensively utilized in various industrial applications. However, assessing the wear behaviour of the HEA coatings through conventional methods remains challenging and time-consuming due to the complexity of the HEA structures. In this study, a novel methodology has been proposed for predicting the wear behaviour of HEA coatings using Machine Learning (ML) algorithms such as Support Vector Machine (SVM), Linear Regression (LR), Gaussian Process Regression (GPR), Least Absolute Shrinkage and Selection Operator (LASSO), Bagging Regression (BR), Gradient Boosting Regression Tree (GBRT), and Robust regressions (RR). The analysis integrates of 75 combinations of HEA coatings with processing parameters and wear test results from peer-reviewed journals for model training and validation. Among the ML models utilized, the GBRT model was found to be more effective in predicting wear rate and Coefficient of Friction (COF) with the highest correlation coefficient of R 2 value of 0.95 ∼ 0.97 with minimal errors. The optimum model is used to predict the unknown wear properties of HEA coatings from the conducted experiments and validate the results, making ML a crucial resource for engineers in the materials sector.