Automatic method for spectral pattern association with characteristic frequencies

Marcin Firla (1) , Timothée Gerber (1) , Pascal Bellemain (2) , Nadine Martin (1)

1 GIPSA-SAIGA - GIPSA - Signal et Automatique pour la surveillance, le diagnostic et la biomécanique
2 GIPSA-Services - GIPSA-Services

This paper proposes an advanced signal-processing technique to improve the condition monitoring of rotating machinery. The proposed method employs the results of a blind spectrum interpretation including harmonic and sideband series detection. The contribution of this paper is an algorithm for automatic association of harmonic and sideband series with the characteristic fault frequencies listed in the kinematic configuration of the monitored system. The proposed algorithm is efficient in inspection of real-world signals, which contain a vast number of detected spectral components. The proposed approach has the advantage of taking into account a possible slip of the rolling-element bearings. The performance of the proposed algorithm is illustrated on real-world data by investigating a shaft problem of an industrial wind turbine high-speed shaft.

Mots clés: wind shaft turbine diagnosis; characteristic fault frequencies; condition monitoring

Domaines: Traitement du signal et de l'image [eess.SP]

Marcin Firla, Timothée Gerber, Pascal Bellemain, Nadine Martin. Automatic method for spectral pattern association with characteristic frequencies. Diagnostyka , 2015, 16 (4), pp.77-84. ⟨hal-01245492⟩