Preprints
https://doi.org/10.5194/wes-2025-131
https://doi.org/10.5194/wes-2025-131
07 Aug 2025
 | 07 Aug 2025
Status: this preprint is currently under review for the journal WES.

Failure classification of wind turbine operational conditions using hybrid machine learning

Marcela Rodrigues Machado, Amanda Aryda Silva Rodrigues de Sousa, Jefferson da Silva Coelho, and Rafael de Oliveira Teloli

Abstract. Wind turbines are complex electromechanical systems that require continuous monitoring to ensure operational efficiency, reduce maintenance costs, and prevent critical failures. Machine learning has shown great promise in structural health monitoring (SHM) by enabling automated fault detection through data-driven approaches. However, challenges remain in adapting SHM methods to complex environmental conditions while maintaining reliable fault detection and classification. This work proposes a hybrid model that combines supervised and unsupervised learning techniques for classifying operational failures in wind turbines. The proposed framework integrates multiphysics data, combining structural and environmental information, to monitor four distinct operational states. The approach begins with the analysis of sensor signals and the extraction of descriptive features that capture the dynamic behaviour of the turbine. The k-means algorithm is applied to label and cluster the dataset, while feature and sensor selection are performed using canonical correlation analysis to rank the most informative variables. A novel relative change damage index is introduced to normalise and scale features based on their relative variability, enhancing the accuracy of clustering and fault classification. Classification is conducted using six different machine learning algorithms. Experimental results demonstrate strong performance in both binary and multiclass tasks, including the detection of pitch drive faults and the accurate identification of rotor icing and aerodynamic imbalance. The model achieved up to 100 % classification accuracy, highlighting its effectiveness in diagnosing wind turbine conditions and improving the overall reliability and operational safety of these systems.

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 paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.
Share
Marcela Rodrigues Machado, Amanda Aryda Silva Rodrigues de Sousa, Jefferson da Silva Coelho, and Rafael de Oliveira Teloli

Status: open (until 11 Sep 2025)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • RC1: 'Comment on wes-2025-131', Anonymous Referee #1, 26 Aug 2025 reply
  • RC2: 'Comment on wes-2025-131', Anonymous Referee #2, 27 Aug 2025 reply
Marcela Rodrigues Machado, Amanda Aryda Silva Rodrigues de Sousa, Jefferson da Silva Coelho, and Rafael de Oliveira Teloli

Model code and software

PyMLDA - Machine Learning for Damage Assessment M. R. Machado, J. S. Coelho, and A. A. de Sousa https://github.com/mromarcela/PyMLDA

Marcela Rodrigues Machado, Amanda Aryda Silva Rodrigues de Sousa, Jefferson da Silva Coelho, and Rafael de Oliveira Teloli

Viewed

Total article views: 236 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
197 26 13 236 8 9
  • HTML: 197
  • PDF: 26
  • XML: 13
  • Total: 236
  • BibTeX: 8
  • EndNote: 9
Views and downloads (calculated since 07 Aug 2025)
Cumulative views and downloads (calculated since 07 Aug 2025)

Viewed (geographical distribution)

Total article views: 235 (including HTML, PDF, and XML) Thereof 235 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 28 Aug 2025
Download
Short summary
We have developed a method using artificial intelligence to detect and classify faults in wind turbines before major damage occurs. By analysing data from multiple sensors, it can identify issues even under changing weather conditions, such as temperature and wind. This improves reliability, reduces downtime, and lowers maintenance costs, supporting cleaner and more affordable energy through stable production.
Share
Altmetrics