The Effect of Noise onto HHT Features for Face Milling Cutter Condition Monitoring

Abstract

In face milling process of train wheel, cutter is one of the most important part and this part should be monitored from failure phenomena for improving final products of train wheel during machining process. One of the best ways for face milling tool condition monitoring is by analyzing signal. However, noise usually contaminates the measured signals during measurement using any sensors. This study presents the effect of noise on the Hilbert-Huang transform features for face milling condition monitoring by mean analyzing the synthetic vibration signals. First, noisy synthetic signals were created, then separates them by Empirical Mode Decomposition (EMD) to be intrinsic mode decompositions (IMFs). Second, the Hilbert-Huang spectra were generated and then compared to HHT baseline spectrum. The result showed that the noise disturbed the HHT spectrum. Without filtering signal, the face milling cutter condition phenomenon was difficult to be revealed by HHT.