Study of using cutting chip color to the tool wear prediction

Abstract

AbstractIn this study, the correlation between chip surface chromaticity and wear of cutting tools is established through experiments, and a system for judging and predicting tool wear by observing chip color is proposed. At present, the life prediction of cutting tools is indirectly measured and predicted by using vibration and current. In this study, chip color change is used to predict tool wear, and back-propagation Artificial Neural Networks (ANN) is used to predict and verify. The average error percentage between the predicted value and the actual value of tool wear is only 1.73% and 1.66%, respectively, which was confirmed by cutting test and verification experiments. This study uses Taylor’s tool life model and chip color to analyze, and after repeated tests and experimental analysis, the average error of repeatability is 4.5%. In the verification of stainless steel cutting hard-cutting materials, the equipment accuracy is between 0.5 and 3.0 color difference values of grade 2 to 3. Therefore, the measurement and model establishment of the system can accurately and quickly predict tool wear. In prediction experiment and analysis, the back neural network is used for test, the maximum error ranges are 0.0012 mm and 0.0097 mm, the mean error percentages are only 1.73% and 1.66%.