Comparison of signal features from time and frequency domain for chatter detection

Abstract

This paper presents the results of the research into the applicability of the features derived by statistical analysis of signals from the machining process in the recognition of occurrence and evolving of chatter vibration. Selected set of statistical features were discussed, 3 of them in the time domain, an one in the frequency domain. Effects of these signal features were discussed through analysis of the results of two experiments with machining. It was shown that standard deviation and the skewness of the force or acceleration signal have some desirable properties in recognition of the moment when chatter vibrations occur.