Rolling bearings are widely used in rotating machinery and their fault is one of the most common causes of industrial machinery failure. Damage identification of roller bearings has been deeply developed to detect faults using vibration-based signal processing. There exist different signal processing techniques to decompose a signal and extract informative features such as EMD and Wavelet transform. EMD is a method for decomposing a multi- component signal into several elementary Intrinsic Mode Functions (IMFs) and has been widely applied to fault diagnosis of rotating machines. However, there are some drawbacks such as stopping criterion for sifting process, mode mixing and border effect problem. Ensemble empirical mode decomposition (EEMD) is a newly developed noise assisted method to solve mode mixing problem exists in empirical mode decomposition (EMD) method. Since the white noise is added throughout the entire signal decomposition process, mode mixing is effectively eliminated. However, there is still a great challenge: identifying two effective parameters (the amplitude of added noise and the number of ensemble trials) which may affect the performance of EEMD. Using low amplitude (relative to the signal), mode mixing cannot be prevented. On the other hand, too large amplitude achieves some redundant IMFs. Although some algorithm or values have been proposed, there is no robust guide to select optimal amplitude yet, especially for early damage detection (very small defects). In this study a reliable method is investigated to determine suitable amplitude and numerous real vibration signals (various operating conditions and two damage locations) are analysed to verify effectiveness and robustness of the proposed method. Vibration signals for healthy and defective bearings were acquired using the test rig assembled by Dynamics & Identification Research Group (DIRG) at Department of Mechanical and Aerospace Engineering, Politecnico di Torino.

Fault diagnosis of roller bearings using ensemble empirical mode decomposition (EEMD) and support vector machine (SVM) / TABRIZI ZARRINGHABAEI, ALI AKBAR; Garibaldi, Luigi; Marchesiello, Stefano; Fasana, Alessandro. - ELETTRONICO. - (2014), pp. 1-12. (Intervento presentato al convegno XIXth SYMPOSIUM Vibrations, shocks and noise tenutosi a Aix en Provence, France nel JUNE 17-19, 2014).

Fault diagnosis of roller bearings using ensemble empirical mode decomposition (EEMD) and support vector machine (SVM)

TABRIZI ZARRINGHABAEI, ALI AKBAR;GARIBALDI, Luigi;MARCHESIELLO, STEFANO;FASANA, ALESSANDRO
2014

Abstract

Rolling bearings are widely used in rotating machinery and their fault is one of the most common causes of industrial machinery failure. Damage identification of roller bearings has been deeply developed to detect faults using vibration-based signal processing. There exist different signal processing techniques to decompose a signal and extract informative features such as EMD and Wavelet transform. EMD is a method for decomposing a multi- component signal into several elementary Intrinsic Mode Functions (IMFs) and has been widely applied to fault diagnosis of rotating machines. However, there are some drawbacks such as stopping criterion for sifting process, mode mixing and border effect problem. Ensemble empirical mode decomposition (EEMD) is a newly developed noise assisted method to solve mode mixing problem exists in empirical mode decomposition (EMD) method. Since the white noise is added throughout the entire signal decomposition process, mode mixing is effectively eliminated. However, there is still a great challenge: identifying two effective parameters (the amplitude of added noise and the number of ensemble trials) which may affect the performance of EEMD. Using low amplitude (relative to the signal), mode mixing cannot be prevented. On the other hand, too large amplitude achieves some redundant IMFs. Although some algorithm or values have been proposed, there is no robust guide to select optimal amplitude yet, especially for early damage detection (very small defects). In this study a reliable method is investigated to determine suitable amplitude and numerous real vibration signals (various operating conditions and two damage locations) are analysed to verify effectiveness and robustness of the proposed method. Vibration signals for healthy and defective bearings were acquired using the test rig assembled by Dynamics & Identification Research Group (DIRG) at Department of Mechanical and Aerospace Engineering, Politecnico di Torino.
File in questo prodotto:
File Dimensione Formato  
Manuscript_VISHNO2014.pdf

accesso aperto

Tipologia: 2. Post-print / Author's Accepted Manuscript
Licenza: PUBBLICO - Tutti i diritti riservati
Dimensione 1.83 MB
Formato Adobe PDF
1.83 MB Adobe PDF Visualizza/Apri
Pubblicazioni consigliate

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11583/2555539
 Attenzione

Attenzione! I dati visualizzati non sono stati sottoposti a validazione da parte dell'ateneo