A Digital Twin Based on OpenFAST Linearizations for Real-Time Load and Fatigue Estimation of Land-Based Turbines

Emmanuel Branlard, Jason Jonkman, Scott Dana, Paula Doubrawa Moreira

Research output: Contribution to journalArticlepeer-review

25 Scopus Citations

Abstract

Monitoring a wind turbine requires intensive instrumentation, which would be too cost-prohibitive to deploy to an entire wind plant. This work presents a technique that uses readily available measurements to estimate signals that would otherwise require additional instrumentation. This study presents a digital twin concept with a focus on estimating wind speed, thrust, torque, tower-top position, and loads in the tower using supervisory control and data acquisition (SCADA) measurements. The model combines a linear state-space model obtained using OpenFAST linearizations, a wind speed estimator, and a Kalman filter algorithm that integrates measurements with the state model to perform state estimations. The measurements are: top acceleration, generator torque, pitch, and rotational speed. The article extends previous work that derived the linear state-space model using a different method. The new implementation, based on OpenFAST linearization capability, allows for a systematic extension of the method to more states, inputs, outputs, and to the offshore environment. Results from the two methods are compared, and the validation is made with additional measurements using the GE 1.5-MW turbine located at the National Renewable Energy Laboratory test site. Real-time damage equivalent loads of the tower bottom moment are estimated with an average accuracy of approximately 10%. Overall, the results from this proof of concept are encouraging, and further application of the model will be considered.

Original languageAmerican English
Article numberArticle No. 022030
Number of pages11
JournalJournal of Physics: Conference Series
Volume1618
Issue number2
DOIs
StatePublished - 22 Sep 2020
EventScience of Making Torque from Wind 2020, TORQUE 2020 - Virtual, Online, Netherlands
Duration: 28 Sep 20202 Oct 2020

Bibliographical note

Publisher Copyright:
© 2020 Published under licence by IOP Publishing Ltd.

NREL Publication Number

  • NREL/JA-5000-76854

Keywords

  • digital twin
  • fatigue
  • Kalman filter
  • loads estimation

Fingerprint

Dive into the research topics of 'A Digital Twin Based on OpenFAST Linearizations for Real-Time Load and Fatigue Estimation of Land-Based Turbines'. Together they form a unique fingerprint.

Cite this