Investigation of a single-layer perceptron neural network to tool wear inception in a metal turning process

Implementation of neural networks to integrate sensor signals in the cutting tool condition monitoring (TCM) problem has been widely pursued, but most of these methods have either been complicated or required detailed sensor signal pre-processing. The authors present a multi-sensor integration method by way of a perceptron neural network to the TCM problem. Three triaxial sensor signals, namely the static cutting force, dynamic cutting force and the vibration signature were used together with the three condition parameters. Successful classification close to 90% was achieved.