A multi-sensor integration method of signals in a metal cutting operation via application of multilayer perceptron neural networks

The potential application of neural networks in manufacturing scenarios is increasingly becoming feasible. Typical of such manufacturing scenario is the integration of metal cutting sensor signals in pursuance of reliable tool condition monitoring (TCM) system. Successful application of this method of sensor integration could save downtime and costs, that would otherwise not have been realised through traditional tool changing philosophies. Unfortunately, the network algorithms used have complicated, requiring detailed sensor signal pre-processing. Partly as a consequence, developed systems have found very limited applications to-date. In this paper, the authors present a simple sensor fusion method via the neural networks approach to the TCM problem. Turning tests were conducted from which the static cutting force, dynamic cutting force and the vibration signature were recorded. The obtained data was used to investigate the classification capability of simple multilayer perceptron (MLP) neural network architectures to the detection of tool wear. Obtained results showed classification accuracy of well over 90%