Modelling, prediction and analysis of surface roughness in turning process with carbide tool when cutting steel C38 using artificial neural network

Farid Boukezzi (1) , Rachid Noureddine (2) , Ali Benamar (3) , Farid Noureddine (4)

1 Université de Saïda Dr Moulay Tahar
2 Université d'Oran 2 Mohamed Ben Ahmed [Oran]
3 ENPO-MA - École Nationale Polytechnique d'Oran Maurice Audin
4 LGP - Laboratoire Génie de Production

Surface roughness is a very important measurement in machining process and a determining factor describing the quality of machined surface. This research aims to analyse the effect of cutting parameters [cutting speed(v), feed rate (f) and depth of cut (d)] on the surface roughness in turning process. For that purpose, an artificial neural network (ANN) model was built to predict and simulate the surface roughness. The ANN model shows a good correlation between the predicted and the experimental surface roughness values, which indicates its validity and accuracy. A set of 27 experimental data on steel C38 using carbide P20 tool have been conducted in this study.

Mots clés: Prediction; Surface roughness; Artificial neural network; Turning; Cutting parameters; ANN; Modelling; Simulation

Domaines: Mécanique des matériaux [physics.class-ph]

Farid Boukezzi, Rachid Noureddine, Ali Benamar, Farid Noureddine. Modelling, prediction and analysis of surface roughness in turning process with carbide tool when cutting steel C38 using artificial neural network. International Journal of Industrial and Systems Engineering, 2017, 26 (4), pp.567-583. ⟨10.1504/IJISE.2017.085227⟩. ⟨hal-02135757⟩