Ball bearings can be affected by several damage typologies. Surface flaws on inner and outer races or on rolling elements are the main causes of failure. The passing of a rolling element upon a localised defect generates a wide band impulse: during machine running, this particular phenomenon repeats itself at the fault characteristic frequencies, which depend on the bearing geometry. The present work shows the results obtained by the application of the main MATLAB Neural Networks to experimental parameters extracted from the casing of the ball bearing of a test machine in operating condition. The analysed bearings were affected by the above mentioned damages, artificially created by electric erosion. A comparison between the results obtained by the application of different network architectures is reported
Ball Bearing Diagnostics Using Neural Networks / Giuliani, G; Rubini, Riccardo; Maggiore, A.. - STAMPA. - (1998), pp. 767-776. (Intervento presentato al convegno International Conference Acoustical and Vibratory Surveillance Methods and Diagnostic Techniques tenutosi a Senlis nel 13-15 ottobre 1998).
Ball Bearing Diagnostics Using Neural Networks
RUBINI, Riccardo;
1998
Abstract
Ball bearings can be affected by several damage typologies. Surface flaws on inner and outer races or on rolling elements are the main causes of failure. The passing of a rolling element upon a localised defect generates a wide band impulse: during machine running, this particular phenomenon repeats itself at the fault characteristic frequencies, which depend on the bearing geometry. The present work shows the results obtained by the application of the main MATLAB Neural Networks to experimental parameters extracted from the casing of the ball bearing of a test machine in operating condition. The analysed bearings were affected by the above mentioned damages, artificially created by electric erosion. A comparison between the results obtained by the application of different network architectures is reported
Pubblicazioni consigliate
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris
