WAVELET TRANSFORM PACKET FOR SIGNAL ANALYSIS USING STANDARD DEVIATION AND SIGNAL SPECTRUM

Authors

  • Nikola Vorkapic Faculty of Mechanical Enginnering
  • Branko Kokotović University of Belgrade, Faculty of Mechanical Enginering
  • Saša Živanović University of Belgrade, Faculty of Mechanical Enginering

Keywords:

chatter vibration, wavelets, spectral analysis, standard deviation

Abstract

The paper presents the research results and the possibilities of applying the standard deviation and amplitude spectrum for chatter vibration detection using the wavelet transform packet as a tool for signal decomposition that was recorded during machining. An analysis of the signal spectrum was additionally performed, which is represented by a spectrogram. Two experiments were performed, where the cutting force signal and acceleration signal, recorded during the machining processes, were analyzed. The recorded signals were decomposed using the wavelet transform packet with the mother wavelet 'db4', and then the standard deviation in the time domain and the amplitude spectrum in the frequency domain were calculated for each level of the decomposed signal. The application of wavelets in this study is intended to enable a better assessment of the quality of standard deviation and amplitude spectrum in chatter vibration detection.

Downloads

Published

2024-04-17

Issue

Section

Articles