Numéro de la revue: Volume 29 , Numéro 2
Auteurs: Fakhreddine Khrouf *
* Mechanics Laboratory, Faculty of Technology Sciences, Mentouri Constantine 1 University, Chaab Ersas Campus, 25000 Constantine, Algeria. https://orcid.org/0000-0002-9884-0803
ABSTRACT
The paper proposes the use of machine learning models to predict the cutting properties when turning EN-AW-1350 aluminum alloy under dry cutting conditions. cutting speeds (m/min), feed rates (mm/rev) and depths of cut (mm) are the main input control parameters selected for the present investigation. The established models can predict the surface roughness values (Ra) during the aluminum hard turning operation, which can guide the direction of experiments and eliminate the need for time-consuming traditional experimental procedures. The prediction performance of six regression methods (Decision Tree (DT), linear regression (LR), eXtreme Gradient Boosting (XGBoost), AdaBoost, Support Vector Machine (SVM) and Random Forest (RF)) were evaluated with a test set/training set ratio of 8 to 2. Among the six regression methods, XGBoost had the best prediction effect on the surface roughness of EN AW-1350 aluminum alloy with a low mean root mean square error (RMSE) and coefficient of determination (R2) close to 1.
Keywords:
Aluminum alloy, Machine learning, Turning, Surface roughness, Regression methods.