Statistical tool breakage detection schemes based on vibration signals in NC milling

We develop a vibration sensor-based tool breakage detection system for NC m illing operations. The system obtains the time-domain vibration signal from the sensor attached on the spindle bracket of our CNC machine and declares tool failures through the on-line monitoring schemes. For the on-line dete ction, our approach is to use the statistical process control methods where control limits or thresholds are automatically calculated independently of cutting conditions. The main thrust of this paper is to compare the perfor mance of the proposed statistical process monitoring methods including the X-bar control scheme, the exponentially weighted moving average (EWMA) sche me, and the adaptive EWMA scheme. The performance of the control schemes ar e compared in terms of the type I and II errors calculated from the experim ent data. (C) 1999 Elsevier Science Ltd. All rights reserved.