On-line metal cutting tool condition monitoring. II: tool-state classification using multi-layer perceptron neural networks

This paper outlines a neural networks based modular tool condition monitori ng system for cutting tool-state classification. Test cuts were conducted o n EN24 alloy steel using P15 and P25 coated cemented carbide inserts and on -line cutting forces and vibration data acquired. Simultaneously the wear l engths on the cutting edges were measured, and these together with the proc essed data were fed to a neural network trained to distinguish tool-state. The first part of the investigation concentrated on tool-state classificati on using a single wear indicator and progressing to two wear indicators. Th e developed system was tested for a variety of cutting conditions and its a bility to distinguish changes in tooling material and cutting conditions fr om those arising from tool wear was assessed. The system was found to be ca pable of accurate tool state classification in excess of 90% accuracy but d eteriorated when the cutting conditions were significantly changed. (C) 200 0 Elsevier Science Ltd. All rights reserved.