Advancing chatter detection: Harnessing the strength of wavelet synchrosqueezing transform and Hilbert-Huang transform techniques

In the manufacturing process, chatter detection is essential to preserving product quality, minimising tool wear, and ensuring efficient productivity. Conventional chatter detection methods often lack the precision required to accurately capture chatter frequencies, which motivates research into advanced signal processing approaches. This paper proposes a wavelet-Hilbert technique (WHT) to get over this limitation of the conventional method. The integration of wavelet synchrosqueezing transform (WSST) and Hilbert-Huang transform (HHT) methods strengthens the robustness of chatter detection algorithms, allowing them to perform effectively across a range of machining conditions. It employs a synchrosqueezing process that increases the time frequency localization, providing the signal component with a clearer representation and increasing detection accuracy. Its integrating nature, which enables comprehensive analysis and effective chatter detection, makes it a novel approach. The force and acceleration signals were used in a comparative test. The comparison analysis demonstrates that signals with lower computing complexity (acceleration signals) are more appropriate. Subsequently, further testing and the collection of acceleration signals were carried out to fully validate the proposed method. The Renyi entropy's value was ascertained. The proposed method offers a higher-resolution TFR and an average Renyi entropy value of 12.3 in comparison to the conventional method's fuzzy representation and entropy value of 15.1.