Tool wear modelling through regression analysis and intelligent methods for nickel base alloy machining

Tool wear measurement data from turning of Inconel 718 aircraft engine components were processed by regression analysis (RA) and artificial neural network (ANN) paradigms, aiming at the on-line prediction of tool wear development. A four-constant empirical model was derived to predict flank wear as a function of the cutting time and cutting speed. These results were compared with the ones obtained from the ANN. The accuracy of the ANN prediction is better than the RA especially when a limited number of data are known. However, ANN required a considerably longer times in the selection of the best network configuration.