Application of artificial neural networks in prediction tool life of PVD coated carbide when end milling of TI6aL4v alloy (original) (raw)

2012, International Journal of Geomechanics

Nowadays, the application of artificial neural networks (ANN) is often utilized in solving numerous problems in machining processes. There has been evidence of the significance of a tool life prediction of coated and uncoated cutting tools. The current study aims at applying ANN in the prediction of the tool life of PVD cutting tools using low experimental data sets. It used a feed forward back propagation neural network with a Levenberg-Marquard (L-M) training algorithm is used in modeling the tool life of a PVD insert cutting tool when end milling of Ti6Al4V under dry cutting conditions. One hundred and ten (110) models were designed, trained and tested using Matlab neural network tool box. Based on the same experimental data, a regression model (RM) has been constructed employing SPSS software, and based on the mean square error of ANN and RM models, the two models were compared. The findings revealed that the ANN model resulted into minimum mean square error compared with RM mod...