Tool wear prediction from acoustic emission and surface characteristics via an artificial neural network

We examine the application of an artificial neural network to classificatio n of tool wear states in face milling. The input features were derived from measurements of acoustic emission during machining and topography of the m achined surfaces. Five input features were applied to the back-propagating neural network to predict a wear state of light, medium or heavy wear. We p resent results from milling experiments with multi- and single-point cuttin g and compare the neural network predictions with observed cutting insert w ear states. (C) 1999 Academic Press.