Development of a machine learning model for wind turbine fatigue and ultimate loads based on static loads

T. Barlas*, T. Göçmen, R. Riva

*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingArticle in proceedingsResearchpeer-review

31 Downloads (Pure)

Abstract

A machine learning model is trained on HAWC2 time-domain aeroelastic load simulations in order to provide a load surrogate model from static rotor loads input to lifetime wind turbine design loads, to be utilized in fast design loads evaluation for design optimization. The simulations are performed on the IEA-3.4-130-RWT with a range of design variations for tip-speed-ratio, pitch-ramp settings, and blade length scaling. The surrogate model is shown to provide accurate predictions of lifetime blade and turbine ultimate and fatigue loads for design variations within the training design space.
Original languageEnglish
Title of host publicationThe Science of Making Torque from Wind (TORQUE 2024): Modeling and simulation technology
Number of pages10
PublisherIOP Publishing
Publication date2024
Article number052009
DOIs
Publication statusPublished - 2024
EventThe Science of Making Torque from Wind (TORQUE 2024) - Florence, Italy
Duration: 29 May 202431 May 2024

Conference

ConferenceThe Science of Making Torque from Wind (TORQUE 2024)
Country/TerritoryItaly
CityFlorence
Period29/05/202431/05/2024
SeriesJournal of Physics: Conference Series
Number5
Volume2767
ISSN1742-6588

Keywords

  • Load surrogate
  • Design loads
  • Machine learning

Fingerprint

Dive into the research topics of 'Development of a machine learning model for wind turbine fatigue and ultimate loads based on static loads'. Together they form a unique fingerprint.

Cite this