555

Wind turbine anomaly detection using a support vector machine optimized with a Bayesian algorithm

D. Coronel(1), C. Guevara(2) and M. Santos(3)

1. Computer Science Faculty, University Complutense of Madrid, Spain

2. Quantitative Methods Department. CUNEF University, Spain

3. Institute of Knowledge Technology, Complutense University of Madrid, Spain

2026-02-15

Wind energy has experienced significant growth inrecent years; however, it still faces challenges in operation and maintenance, which impact energy efficiency and lead to high costs. This study proposes an anomaly detection model for wind turbines based on a support vector machine (SVM), optimized using Bayesian search. The model was trained using vibration data from the Gearbox Reliability Collaborative (GRC) project of the National Renewable Energy Laboratory (NREL), specifically from the High-Speed Shaft Upwind Bearing Radial and High-Speed Shaft Downwind Bearing Radial sensors, with 40,000 records evenly distributed between normal and anomalous conditions. The proposed model achieved an overall fault detection accuracy of 78.95% and 78.50% for the respective sensor data. Bayesian optimization facilitated the fine-tuning of the hyperparameters in the classification technique, enhancing the model's anomaly detection capability. Furthermore, the use of vibration data enabled the identification of critical fault patterns in turbine operation, contributing to the improvement of wind turbine efficiency and reliability.

Key words: Support Vector Machines, Anomalies, Wind Turbines, Bayesian algorithm.

[1] T. Matsui, K. Yamamoto, and J. Ogata, “Anomaly detection for wind turbine damaged due to lightning strike,” Electric Power Systems Research, vol. 209, Aug. 2022, doi: 10.1016/j.epsr.2022.107918.

[2] R. Pandit, D. Infield, and M. Santos, “Accounting for Environmental Conditions in Data-Driven Wind Turbine Power Models,” IEEE Trans Sustain Energy, vol. 14, pp. 168–177, Jan. 2023, doi: 10.1109/TSTE.2022.3204453.

[3] Global Wind Energy Council, “Global Wind Report 2024.” Accessed: Dec. 19, 2024. [Online]. Available: https://gwec.net/global-wind-report-2024/

[4] J. Chen, J. Li, W. Chen, Y. Wang, and T. Jiang, “Anomaly detection for wind turbines based on the reconstruction of condition parameters using stacked denoising autoencoders,” Renew Energy, vol. 147, pp. 1469–1480, Mar. 2020, doi: 10.1016/J.RENENE.2019.09.041.

[5] E. S. Miele, F. Bonacina, and A. Corsini, “Deep anomaly detection in horizontal axis wind turbines using Graph Convolutional Autoencoders for Multivariate Time series,” Energy and AI, vol. 8, May 2022, doi: doi.org/10.1016/j.egyai.2022.100145.

[6] R. Pandit, M. Santos, and J. E. Sierra-García, “Comparative analysis of novel data-driven techniques for remaining useful life estimation of wind turbine high-speed shaft bearings,” Energy Sci Eng, Oct. 2024, doi: 10.1002/ese3.1911.

[7] M. Serrano-Antoñanazas, J. E. Sierra-Garcia, M. Santos, and M. Tomas-Rodríguez, “Identification of Vibration Modes in Floating Offshore Wind Turbines,” J Mar Sci Eng, vol. 11, Oct. 2023, doi: 10.3390/jmse11101893.

[8] H. Chen, H. Liu, X. Chu, Q. Liu, and D. Xue, “Anomaly detection and critical SCADA parameters identification for wind turbines based on LSTM-AE neural network,” Renew Energy, vol. 172, pp. 829–840, Jul. 2021, doi: 10.1016/j.renene.2021.03.078.

[9] M.-H. ; Wang et al., “Fault Detection of Wind Turbine Blades Using Multi-Channel CNN,” Sustainability 2022, Vol. 14, Page 1781, vol. 14, no. 3, p. 1781, Feb. 2022, doi: 10.3390/SU14031781.

[10] J. Liu, G. Yang, X. Li, Q. Wang, Y. He, and X. Yang, “Wind turbine anomaly detection based on SCADA: A deep autoencoder enhanced by fault instances,” ISA Trans, vol. 139, pp. 586–605, Aug. 2023, doi: 10.1016/J.ISATRA.2023.03.045.

[11] J. Cervantes, F. Garcia-Lamont, L. Rodríguez-Mazahua, and A. Lopez, “A comprehensive survey on support vector machine classification: Applications, challenges and trends,” Neurocomputing, vol. 408, pp. 189–215, Sep. 2020, doi: 10.1016/j.neucom.2019.10.118.

[12] S. M. Lundberg, G. G. Erion, and S.-I. Lee, “An interpretable model for landslide susceptibility assessment based on Optuna hyperparameter optimization and Random Forest,” Geomatics, Natural Hazards and Risk, vol. 15, no. 1, Apr. 2024, doi: 10.1080/19475705.2024.2347421.

[13] V. R. Joseph, “Optimal ratio for data splitting,” Stat Anal Data Min, vol. 15, pp. 531–538, Aug. 2022, doi: 10.1002/sam.11583.

[14] S. Cuéllar, M. Santos, F. Alonso, E. Fabregas, and G. Farias, “Explainable anomaly detection in spacecraft telemetry,” Eng Appl Artif Intell, vol. 133, Jul. 2024, doi: 10.1016/j.engappai.2024.108083.