A study is presented in this paper able to identify and to quantify the effects of sensor characteristics and of data processing aspects on the uncertainty of the features used for Condition Monitoring (CM) applications, based on a hybrid approach. The precision of the sensors and the modalities to perform the FFT and obtain the related features are evaluated, to improve the coherence of information deriving from the data obtained by means of a physics-based model and that gained from the experiments. Validation of the features related to both classes is expected to improve the fusion process of data and the accuracy of prognostic algorithms.
Validation of signal processing techniques for vibration measurements
Giulio D’Emilia
Membro del Collaboration Group
;Antonella GaspariMembro del Collaboration Group
;Emanuela NataleMembro del Collaboration Group
2017-01-01
Abstract
A study is presented in this paper able to identify and to quantify the effects of sensor characteristics and of data processing aspects on the uncertainty of the features used for Condition Monitoring (CM) applications, based on a hybrid approach. The precision of the sensors and the modalities to perform the FFT and obtain the related features are evaluated, to improve the coherence of information deriving from the data obtained by means of a physics-based model and that gained from the experiments. Validation of the features related to both classes is expected to improve the fusion process of data and the accuracy of prognostic algorithms.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
