@inproceedings{01JRW7GM7WJY0KJSHS5E1WAWPB,
  abstract     = {{Accurate modeling of wind turbine dynamics is essential, as it enables engineers to design optimized control systems that enhance the performance and reliability of wind turbines. This study demonstrates how a hybrid model, integrating physics-based equations and a data-driven neural network, outperforms both low-fidelity physics models and purely data-driven approaches in estimating rotational speed with higher accuracy. By leveraging fundamental physical principles to capture key aerodynamic and mechanical behaviors and incorporating machine learning to estimate effective wind speed, the hybrid approach addresses the limitations of traditional methods. Purely physics-based models often struggle with uncertainties and variability in real-world conditions, while data-driven models lack generalizability and fail to incorporate physical constraints. In contrast, the hybrid model combines the strengths of both approaches, ensuring consistency with physical laws while adapting to operational variability. Validation using the high-fidelity OpenFAST simulation environment highlights the potential of the hybrid model not only for accurate dynamic modeling but also as a foundation for future model-based optimal control designs.}},
  articleno    = {{012008}},
  author       = {{Heidari, Jalal and Kayedpour, Nezmin and Vandevelde, Lieven and Crevecoeur, Guillaume}},
  booktitle    = {{WINDEUROPE 2025}},
  issn         = {{1742-6588}},
  language     = {{eng}},
  location     = {{Copenhagen, Denmark}},
  number       = {{1}},
  pages        = {{11}},
  title        = {{Physics-based, data-augmented model for wind turbine control design}},
  url          = {{http://doi.org/10.1088/1742-6596/3025/1/012008}},
  volume       = {{3025}},
  year         = {{2025}},
}

Altmetric

View in Altmetric

Web of Science

Times cited: