EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Wind Turbine Load Reduction Predictive Control Strategy Based on Dynamic Model and Data‐Driven Approach

Bin Wang, Zhiguo Zhao, Yuyi Wang, Jun Lin, Tao Chen, Zixiang Xiong, Wei He

IET Renewable Power Generation · 2026-01 · 20 (1)

DOI: 10.1049/rpg2.70337 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IET Renewable Power Generation, online erschienen am 10.08.2026. Verfasst von Bin Wang und 6 weiteren Autor:innen. Die Arbeit nennt 31 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Natural Science Foundation of Hunan Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

To suppress the aerodynamic and structural loads of large‐scale wind turbines and mitigate modelling errors induced by parameter uncertainties and unmodeled dynamics, this paper proposes a data‐driven corrected dynamic modelling and predictive control strategy for wind turbines operating within a full operational wind speed range of 3–25 m/s. A hybrid modelling framework is established by combining a physics‐based dynamic model with a three‐layer neural network correction module. Specifically, full wind‐speed‐range operational data collected from Bladed simulations are employed to refine the key parameters of the mechanism model and the neural network with a 64‐32‐5 neuron configuration and dual sigmoid/linear activation functions are adopted to compensate for residual dynamic errors, effectively improving the model accuracy under time‐varying wind conditions. On this basis, a data‐driven dynamic model predictive control (DDMPC) strategy is developed for wind turbine load reduction. Meanwhile, a polynomial wind speed prediction model is constructed to fuse Lidar‐measured and power‐estimated wind speed data, so as to improve the wind preview performance of the control system. Simulations are carried out on a 6.25 MW wind turbine covering the entire 3–25 m/s wind speed range. The results show that compared with conventional physics‐based MPC and PID control, the proposed DDMPC reduces the Y‐direction tower‐base bending moment by 13.8% and the Y‐direction blade‐root bending moment by 5.8%. Further field tests on a practical 10 MW wind turbine demonstrate that the DDMPC strategy significantly improves the stability of generator speed and output power under medium‐low and medium‐high wind speed conditions. This work provides an effective technical solution for high‐precision dynamic modelling and load‐optimal control of large‐scale wind turbines, offering promising engineering application value for improving wind turbine structural safety and operational stability.

Förderung

  • National Natural Science Foundation of China
  • Natural Science Foundation of Hunan Province

Zitieren

APA

Wang, B., Zhao, Z., Wang, Y., Lin, J., Chen, T., Xiong, Z., & He, W. (2026). Wind Turbine Load Reduction Predictive Control Strategy Based on Dynamic Model and Data‐Driven Approach. IET Renewable Power Generation, 20(1). https://doi.org/10.1049/rpg2.70337

Vancouver

Wang B, Zhao Z, Wang Y, Lin J, Chen T, Xiong Z, et al. Wind Turbine Load Reduction Predictive Control Strategy Based on Dynamic Model and Data‐Driven Approach. IET Renewable Power Generation. 2026;20(1). doi:10.1049/rpg2.70337

BibTeX
@article{Wang2026,
  author = {Wang, Bin and Zhao, Zhiguo and Wang, Yuyi and Lin, Jun and Chen, Tao and Xiong, Zixiang and He, Wei},
  title = {Wind Turbine Load Reduction Predictive Control Strategy Based on Dynamic Model and Data‐Driven Approach},
  journal = {IET Renewable Power Generation},
  year = {2026},
  volume = {20},
  number = {1},
  doi = {10.1049/rpg2.70337}
}
Kostenlos starten →