Skip to Main content Skip to Navigation
Conference papers

Bearing Fault Detection Using Intrinsic Mode Functions Statistical Information

Abstract : This paper investigates the bearing fault detection using vibration signals. For these mechanical type of fault, the energy in the signal is distributed in the frequency domain which is relatively wide. These particular frequencies linked to the fault occurrence is difficult to isolate and characterise. In the case of small faults (namely incipient faults), this location will be somehow difficult. From this point of view, envelop detection methods are used. This work is based on a decomposition method into frequency bands, the Empirical Mode Decomposition (EMD), coupled with a statistical analysis for the selection of the most relevant mode function and the fault detection evaluation. Thanks to the selection of the most sensitive Intrinsic Mode Function (IMF), the results based on experimental data shows the effectiveness of bearing ball fault detection using the two first statistical moments.
Document type :
Conference papers
Complete list of metadatas
Contributor : Claude Delpha <>
Submitted on : Friday, May 18, 2018 - 10:22:30 AM
Last modification on : Wednesday, September 16, 2020 - 5:45:27 PM



Zahra Mezni, Claude Delpha, Demba Diallo, Ahmed Braham. Bearing Fault Detection Using Intrinsic Mode Functions Statistical Information. IEEE 15th International Multi-Conference on Systems, Signals and Devices (SSD'18), Mar 2018, Hammamet, Tunisia. ⟨10.1109/ssd.2018.8570465⟩. ⟨hal-01795069⟩



Record views