Machine chattering identification based on the fractional-order chaotic synchronization dynamic error

Abstract

The topic of this study is using the fractional-order chaotic synchronization system to identify chattering that CNC machines produce during production. The appearance of chattering indicates instability during the metal milling process which will not only cause abnormal wear on the tool, but can also decrease the precision of the work piece significantly. Thus, identification of chattering has always been a very important research topic. However, previous chattering identification mostly relied on the experience of the operator. Most past studies were based on the energy perspective. When the main frequency in the frequency domain analysis to the existing spindle rotation frequency ratio is a non-integer multiple, then chattering has occurred. We propose a brand new chattering identification method which uses the synchronization error plane centroid in fractional-order chaotic synchronization system to find chattering. Thus, we can simply use position of the chaotic centroid to determine whether or not the current cutting status has chattering. The result of this study shows that the method we proposed can effectively identify chattering and that the identification result is very accurate and useful.