Tool Flank Wear Measurement Using BPN

A reliable and sensitive technique to monitor the tool wear without interrupting the process became a crucial issue for realization of the modern manufacturing concepts without human resources at machining centers. A proper oniine cutting tool condition monitoring system is essential for deciding when change the tool. This paper outlines a neural network based tool condition monitoring system for cutting tool flank wear. The cutting tests were performed on mild steel using HSS tool and on-line cutting forces data was acquired with a cutting force dynamometer. Simultaneously flank wear was measured using a toolmakers microscope and the processed data were fed to a Back Propagation Neural Network (BPNN). The developed system then was tested to predict flank wear for various cutting conditions. The output of the neural network is capable of accurate tool wear prediction for the range it had been trained but the accuracy deteriorated as the cutting conditions were changed significantly.