Volume 11 Issue 2 ( June )

Downtime, poor surface quality, excessive noise, dimension inaccuracy, and disproportionate tool wear are all negative effects of the regenerative chatter phenomena. On obtained audio signals generated during machining of Al-6061-T6, empirical mode decomposition (EMD) and local mean decomposition (LMD) techniques are performed, followed by rapid Fourier transform (FFT). In order to detect and evaluate chatter in its early stages, these decomposed signals are examined to extract tool chatter traits by analysing different statistical indicators such as kurtosis, absolute mean amplitude, impulse index, root mean square value, and so on. The Nakagami probability distribution is used to determine statistical indicator threshold limits. According to the results of the investigation, LMD is a more efficient technique than EMD.

Keywords: Nakagami distribution; tool chatter; EMD; LMD; statistical parameters; non- stationary signal processing.