Preprints
https://doi.org/10.5194/wes-2025-112
https://doi.org/10.5194/wes-2025-112
01 Jul 2025
 | 01 Jul 2025
Status: this preprint is currently under review for the journal WES.

Load Estimation in Onshore Wind Farms Using Surrogate Modeling and Generic Turbine Models

Alexander Mönnig, Ansgar Hahn, Astrid Lampert, and Ulrich Römer

Abstract. This article investigates the development and application of surrogate models, based on slightly adapted generic turbine models, for predicting loads on real-world wind turbines. A small set of aeroelastic simulations provided training data for both Polynomial Chaos expansion and Gaussian Process regression models, which were trained to predict blade loads, tower accelerations, and their respective seed-to-seed variability. To evaluate the practical suitability of these models a case study was performed. Here, the surrogate models were applied to predict blade loads and tower accelerations respectively, using five years of SCADA data from an onshore wind farm. While the models approximated the real-world turbine behavior with a reasonable accuracy, the prediction quality varied across the different turbines in the park and was further influenced by factors such as the turbine's operational years and diurnal patterns suggesting a correlation with the turbulence intensity. Despite some limitations, the findings support the practicality of developing surrogate models for enabling efficient load estimations.

Competing interests: This research was conducted as part of Alexander Mönnig’s Master’s thesis in cooperation with Alterric Deutschland GmbH, where Alexander Mönnig was employed as a working student. Co-author Ansgar Hahn, an employee of Alterric Deutschland GmbH, supported the thesis as an industry supervisor. The company contributed by providing data access and domain expertise. The authors declare no other competing interests.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.
Share
Alexander Mönnig, Ansgar Hahn, Astrid Lampert, and Ulrich Römer

Status: open (until 03 Aug 2025)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on wes-2025-112', Anonymous Referee #1, 22 Jul 2025 reply
  • RC2: 'Comment on wes-2025-112', Anonymous Referee #2, 24 Jul 2025 reply
Alexander Mönnig, Ansgar Hahn, Astrid Lampert, and Ulrich Römer

Data sets

Load Estimation in Onshore Wind Farms Using Surrogate Modelling and Generic Turbine Models Alexander Mönnig, Ulrich Römer https://doi.org/10.5281/zenodo.15380254

Model code and software

alexandermoennig/wind-farm-load-estimation: Wind Farm Load Estimation v1.0.0 Alexander Mönnig, Ulrich Römer https://doi.org/10.5281/zenodo.15446361

Alexander Mönnig, Ansgar Hahn, Astrid Lampert, and Ulrich Römer

Viewed

Total article views: 157 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
126 20 11 157 6 7
  • HTML: 126
  • PDF: 20
  • XML: 11
  • Total: 157
  • BibTeX: 6
  • EndNote: 7
Views and downloads (calculated since 01 Jul 2025)
Cumulative views and downloads (calculated since 01 Jul 2025)

Viewed (geographical distribution)

Total article views: 157 (including HTML, PDF, and XML) Thereof 157 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 24 Jul 2025
Download
Short summary
To enable efficient load estimation in wind farms, e.g. to assess Remaining Useful Life, we investigated the feasibility of using surrogate models based on open-source turbine models. In a case study, we adapted such a model to match turbines from a real German onshore wind farm, ran aeroelastic simulations, and trained surrogate models to reconstruct loads observed in 5 years of SCADA data. Comparing predicted and measured values showed promising accuracy, especially for blade bending moments.
Share
Altmetrics