AStrion data validation of non-stationary wind turbine signals Topic: Condition monitoring (CM) methods and technologies
Résumé
AStrion is an automatic spectrum analyzer software, which proposes a new generic and data-drivenmethod without any a priori information on the measured signals. In order to compute some generalcharacteristics and derive properties of the signal, the first step in this approach is to give some insightinto the nature of the signal. This preanalysis, so called the data validation, contains a number of tests toreveal some properties and characteristics of the data such as the acquisition validity (absence ofsaturation and a posteriori respect of the sampling theorem), the stationarity (or non-stationarity), theperiodicity and the signal-to-noise ratio. Based on these characteristics, the proposed method definesindicators and alarm trigger thresholds, also categorizes the signal into three classes which help to guidethe following spectral analysis. The present paper introduces the four tests of this preanalysis and thesignal categorization rules. Finally, the proposed approach is validated on a set of wind turbine vibrationmeasurements to demonstrate its applicability of a long-term and continuous monitoring on real-worldsignals
Origine : Fichiers produits par l'(les) auteur(s)
Loading...