ANN-based prediction of surface and hole quality in drilling of AISI D2 cold work tool steel

Abstract

This paper focuses on artificial neural network (ANN)-based modeling of surface and hole quality in drilling of AISI D2 cold work tool steel with uncoated titanium nitride (TiN) and titanium aluminum nitride (TiAlN) monolayer- and TiAlN/TiN multilayer-coated-cemented carbide drills. A number of drilling experiments were conducted at all combinations of different cutting speeds (50, 55, 60, and 65 m/min) and feed rates (0.063 and 0.08 mm/rev) to obtain training and testing data. The experimental results showed that the surface roughness (Ra) and roundness error (Re) values were obtained with the TiN monolayer- and TiAlN/TiN multilayer-coated drills, respectively. Using some of the experimental data in training stage, an ANN model was developed. To evaluate the performance of the developed ANN model, ANN predictions were compared with the experimental results. It was found that the determination coefficient values are more than 0.99 for both training and test data. Root mean square error and mean error percentage values were very low. ANN results showed that ANN can be used as an effective modeling technique in accurate prediction of the Ra and Re.