Study of spindle vibration signals for tool breakage monitoring in micro-drilling

Abstract

The study of the spindle vibration signal for micro tool breakage monitoring was reported in this paper. The vibration signal obtained from the three axis accelerometer installed on a fixture connected to the spindle housing were considered as the input signal for the monitoring system which includes the Wavelet transformation of the collected signal, feature selection for creating the features related to the breakage event, and a classifier design for classifying the tool breakage event based on the selected features. The linear discriminate function was developed as the classifier in the system. The effects of the sensor direction and the selected features on the classification rate were discussed as well in this study. In order to collect the vibration signals for training and verifying the system, an experiment was implemented on a micro milling platform along with 600 μm diameter micro drill and 6061 aluminum workpiece. The results show that the spindle vibration signals provides the capability of detecting the micro drill breakage, and the features selected from the Wavelet transformation can improve the reliability of the monitoring system.