@Article{ee.2025.068712,
AUTHOR = {Yang Liu, Lifu Ding, Zhenfan Yu, Tannan Xiao, Qiuyu Lu, Ying Chen, Weihua Wang},
TITLE = {Maximizing Wind Farm Power Output through Site-Specific Wake Model Calibration and Yaw Optimization},
JOURNAL = {Energy Engineering},
VOLUME = {122},
YEAR = {2025},
NUMBER = {11},
PAGES = {4365--4384},
URL = {http://www.techscience.com/energy/v122n11/64225},
ISSN = {1546-0118},
ABSTRACT = {Wake effects in large-scale wind farms significantly reduce energy capture efficiency. Active Wake Control (AWC), particularly through intentional yaw misalignment of upstream turbines, has emerged as a promising strategy to mitigate these losses by redirecting wakes away from downstream turbines. However, the effectiveness of yaw-based AWC is highly dependent on the accuracy of the underlying wake prediction models, which often require site-specific adjustments to reflect local atmospheric conditions and turbine characteristics. This paper presents an integrated, data-driven framework to maximize wind farm power output. The methodology consists of three key stages. First, a practical simulation-assisted matching method is developed to estimate the True North Alignment (TNA) of each turbine using historical Supervisory Control and Data Acquisition (SCADA) data, resolving a common source of operational uncertainty. Second, key wake expansion parameters of the Floris engineering wake model are calibrated using site-specific SCADA power data, tailoring the model to the Jibei Wind Farm in China. Finally, using this calibrated model, the derivative-free solver NOMAD is employed to determine the optimal yaw angle settings for an 11-turbine cluster under various wind conditions. Simulation studies, based on real operational scenarios, demonstrate the effectiveness of the proposed framework. The optimized yaw control strategies achieved total power output gains of up to 5.4% compared to the baseline zero-yaw operation under specific wake-inducing conditions. Crucially, the analysis reveals that using the site-specific calibrated model for optimization yields substantially better results than using a model with generic parameters, providing an additional power gain of up to 1.43% in tested scenarios. These findings underscore the critical importance of TNA estimation and site-specific model calibration for developing effective AWC strategies. The proposed integrated approach provides a robust and practical workflow for designing and pre-validating yaw control settings, offering a valuable tool for enhancing the economic performance of wind farms.},
DOI = {10.32604/ee.2025.068712}
}