Time Localization of Abrupt Changes in Cutting Process using Hilbert Huang Transform

Abstract

Cutting process is extremely dynamical process influenced by different phenomena such as chip formation, dynamical responses and condition of machining system elements. Different phenomena in cutting zone have signatures in different frequency bands in signal acquired during process monitoring. The time localization of signal’s frequency content is very important. An emerging technique for simultaneous analysis of the signal in time and frequency domain that can be used for time localization of frequency is Hilbert Huang Transform (HHT). It is based on empirical mode decomposition (EMD) of the signal into intrinsic mode functions (IMFs) as simple oscillatory modes. IMFs obtained using EMD can be processed using Hilbert Transform and instantaneous frequency of the signal can be computed. This paper gives a methodology for time localization of cutting process stop during intermittent turning. Cutting process stop leads to abrupt changes in acquired signal correlated to certain frequency band. The frequency band related to abrupt changes is localized in time using HHT. The potentials and limitations of HHT application in machining process monitoring are shown.