applicability of machine learning to predict the wear behaivor of ultrahigh- carbon steel

Machine learning (ml) has developed into an effective technique for predicting wear behavior of solid materials. in the present work, the prediction of wear performance of ultrahighcarbon steel under different operating conditions (sliding speed, normal pressure and sliding distance) have been carried out. supervised machine learning algorithms such as linear regression (lr) and random forest (rf) were used with the experimental results as input dataset to predict the wear rate. the two adopted machine learning algorithms are ranked in the order of prediction accuracy: rf and lr, rf has yielded the superior results in r^2 (training and test), mae and rmse among all the constructed models. the findings could aid in the developing of ultrahigh carbon steel with controlled wear qualities, speeding up the development of newfunctional ultrahigh-carbon steel.