Real-time chatter detection and automatic suppression for intelligent spindles based on wavelet packet energy entropy and local outlier factor algorithm
Journal
The International Journal of Advanced Manufacturing Technology
Journal Volume
103
Journal Issue
1-4
Pages
297-309
Date Issued
2019
Author(s)
Abstract
In this research, we proposed a real-time chatter detection and suppression module for intelligent spindle to increase machining efficiency and processing yield. For early detections of chatters, the relative wavelet packet energy entropy with high sensitivity in the high-frequency band and the local outlier factor (LOF) algorithm were utilized as chatter features and classifications, respectively. Based on the pre-obtained three-dimensional stability lobe diagram (SLD) and a LOF-based trained model, the module could real-time monitor and suppress chatter during machining processes. The module was implemented and experimentally examined with the CNC end-milling machine under five different cutting conditions for verifying the capabilities of real-time chatter identifications and suppressions. It was demonstrated that the module could detect the onset of chatter and suppress it by changing the cutting conditions to avoid damages on the surfaces of working piece due to severe chatters.
Type
journal article