Continuous wavelet transform-based method for enhancing estimation of wind turbine blade natural frequencies and damping for machine learning purposes

R. Janeliukstis*

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

Abstract

In the current study, an operational modal analysis is performed on a wind turbine blade with pull-and-release excitation at the tip. The experiments were carried out in large-scale facility of Risoe campus in Technical University of Denmark. Two different blade configurations are compared – blade on the block and a configuration for fatigue testing comprising of blade on the block with mass resonance exciter and seismic masses. A continuous wavelet transform is employed for estimation of modal parameters of the fundamental flapwise bending mode of the blade from acceleration responses. Statistical analysis along with an outlier removal from the extracted modal parameters is carried out to provide clear estimates of the extracted values. These results are compared to the values extracted with other algorithms. The proposed method for modal parameter extraction aims at extraction of numerous observations allowing for statistical decision making and machine learning capabilities.
Original languageEnglish
Article number108897
JournalMeasurement
Volume172
Number of pages20
ISSN0263-2241
DOIs
Publication statusPublished - 2021

Keywords

  • Wind turbine blade
  • Operational mode analysis
  • Modal parameters
  • Continuous wavelet transform
  • Machine learning

Fingerprint Dive into the research topics of 'Continuous wavelet transform-based method for enhancing estimation of wind turbine blade natural frequencies and damping for machine learning purposes'. Together they form a unique fingerprint.

Cite this