Repository logo
 

Applied Mathematical Methods for Vibration-Based Diagnosis of Rotating Machinery Faults

Loading...
Thumbnail Image

Date

2026-06-09

Journal Title

Journal ISSN

Volume Title

Publisher

UNIVERSITE DE ECHAHID CHEIKH LARBI TEBESSI

Abstract

Rolling bearings are essential components of rotating machinery, whose failure can lead to costly production downtime or serious accidents. In this work, two signal processing approaches are compared for bearing fault diagnosis: the Fast Fourier Transform (FFT) and the Hilbert-Huang Transform (HHT). Although the FFT is widely used, it has inherent limitations when applied to non-stationary and nonlinear signals. The HHT, based on Empirical Mode Decomposition (EMD) followed by Hilbert Spectral Analysis (HSA), constitutes a fully adaptive approach requiring no a priori basis. Both methods are applied to vibration signals from the Case Western Reserve University (CWRU) bearing database for an SKF 6205-2RS JEM bearing, under no-load and loaded conditions. Results show that the HHT provides better spectral resolution and more accurate identification of fault-characteristic frequencies, particularly in terms of sideband and harmonic visibility, compared to classical FFT.

Description

Keywords

Fast Fourier Transform (FFT), Hilbert-Huang Transform (HHT), Empirical Mode Decomposition (EMD), Fault diagnosis

Citation