Self-consistent model for active control of wind turbine wakes

Zhaobin Li 1,2, Xiaolei Yang 1,2 · 2025

Read Self-consistent model for active control of wind turbine wakes by Zhaobin Li 1,2, Xiaolei Yang 1,2 (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-12898.txt.

Introduction

Zhaobin Li and Xiaolei Yang built a fast predictive model in 2025 for steering wind turbine wakes. A turbine's wake steals wind from machines behind it. Active wake control wiggles the rotor to help the wake recover sooner. But designing that control needs a model of the coupled physics, and full models are far too slow to design with. Their model asks for only two inputs: a reference wake with no control, and the control plan to test.

Physics-based resolvent study handles the large coherent forms of the wake, an eddy thickness model covers the small-scale turbulence. And a Reynolds stress model ties how both kinds of swing reshape the average flow. Checked against large-eddy models across control methods and wiggle rates. The model forecast the enhanced average wake recovery with only a few percent error, at a small fraction of the model cost.

Catalog record

CREATORZhaobin Li 1,2, Xiaolei Yang 1,2
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE103 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-12898

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Self-consistent model for active control of wind turbine wakes Zhaobin Li 1,2, Xiaolei Yang 1,2 Abstract Active wake control (AWC) has emerged as a promising strategy for enhancing wind turbine wake recovery, but accurately modelling its underlying fluid mechanisms remains challenging. This study presents a computationally efficient wake model that provides end-to-end prediction capability from rotor actuation to wake recovery enhancement by capturing the coupled dynamics of wake meandering and meanflow modification, requiring only two inputs: a reference wake without control and a user-defined AWC strategy. The model combines physics-based resolvent modelling for large-scale coherent structures and an eddy viscosity modelling for small-scale turbulence. A Reynolds stress model is introduced to account for the influence of both coherent and incoherent wake fluctuations, so that the time-averaged wake recovery enhanced by the AWC can be quantitatively predicted. Validation against large-eddy simulations (LES) across various AWC approaches and actuating frequencies demonstrates the model’s predictive capability, accurately capturing AWC-specific and frequency-dependent mean wake recovery with less than 8% error from LES while reducing computational time from thousands of CPU hours to minutes. The efficiency and accuracy of the model makes it a promising tool for practical AWC design and optimization of large-scale wind farms. keywords Authors should not enter keywords on the manuscript, as these must be chosen by the author during the online submission process and will then be added during the typesetting process (see Keyword PDF for the full list). Other classifications will be added at the same time. † † affiliation: 1 The State Key Laboratory of Nonlinear Mechanics, Institute of Mechanics, Chinese Academy of Sciences, Beijing 100190, China 2 School of Engineering Sciences, University of Chinese Academy of Sciences, Beijing 100049, China 1 Introduction Wind farms have emerged as the predominant form of large-scale wind energy development. Within these farms, the proximity of turbines leads to significant aerodynamic interactions, particularly in the form of wake effects ( Vermeer et al. 2003 ; Stevens & Meneveau 2017 ; Meneveau 2024 ) . The wake generated by upstream turbines substantially influences the inflow conditions for downstream turbines, manifesting as reduced wind speeds and increased turbulence intensity ( Porté-Agel et al. 2020 ) , reducing both power generation and structural lifetime of wind turbines. Consequently, understanding and controlling wind turbine wakes are crucial for optimizing the economics and safety of wind farms ( Shapiro et al. 2022 ) and lays the foundation for even higher levels of the electricity grid and market management ( Eguinoa et al. 2021 ) . Fluid mechanics is deeply involved in this research area ( Kheirabadi & Nagamune 2019 ; Abkar et al. 2023 ) . Substantial advancements in wake control have been achieved, and the research evolves from solely static wake steering to including more sophisticated active wake control (AWC) methods ( Meyers et al. 2022 ) . While static wake steering (mainly through yaw misalignment) has demonstrated effectiveness in redirecting wakes away from downstream turbines ( Bastankhah & Porté-Agel 2016 ; Shapiro et al. 2018 ; Howland et al. 2022 ; Meneveau 2024 ) , AWC has emerged as a rather new and promising approach under active development. 1.1 State-of-the-art for AWC modelling AWC enhances wake recovery through well-designed dynamic perturbations of rotor force or position to generate wake pulsing or wake meandering fluctuations ( Houck 2022 ) . Pulsing-type AWCs induce periodic expansion and contraction of the wake as it travels downstream. This wake pattern can be triggered through the dynamic induction control that generates oscillatory thrust forces via collective blade pitch or generator torque adjustments ( Goit & Meyers 2015 ; Munters & Meyers 2017 ) , or through the surge motion of floating offshore wind turbines (FOWTs) ( Messmer et al. 2024 ; Wang et al. 2024 ) . In contrast, meandering-type AWCs create large-scale and oscillatory lateral or helical wake deflections. Exemplary approaches are the dynamic individual pitch control (DIPC) method that triggers helical wake meandering through phase-shifted periodic variation of blade pitch angle ( Frederik et al. 2020 ) , dynamic yaw or sway method ( Munters & Meyers 2018 ; Lin & Porté-Agel 2024 ; Li et al. 2022 ) that induce horizontal wake meandering through periodic variation of yaw angle or horizontal wake position. Due to stronger cross-flow mixing and wake recovery enhancement, the meandering-type AWCs are more efficient than the pulsing-type ones ( Messmer et al. 2024 ) . For this purpose, the present work focuses on meandering-type AWCs and puts emphasis on their fast predictive modelling, which is indispensable for AWC design and optimization. To this end, recent studies on AWC modelling have developed the predictive capability mainly in predicting the optimal frequencies and fluctuation of the wake, but the predictive capability for the final target, i.e., the AWC-induced wake recovery, is still lacking. • Optimal frequency: Wind turbine wakes are convectively unstable and act as amplifiers of upstream perturbations, if the perturbation frequency falls within the wake’s unstable frequency range ( Iungo et al. 2013 ; Mao & Sørensen 2018 ; Gupta & Wan 2019 ) . For this reason, the unstable frequency range of the wake can serve as a good indicator for designing optimal forcing frequency. In this regard, linear stability analysis (LSA) is often applied to the time-averaged wake field without AWC ( Iungo et al. 2013 ; Gupta & Wan 2019 ) . Axisymmetric baseflow assumption is often applied in these analyses, such that the meandering-type fluctuation can be modelled by perturbations with azimuthal wave number \mathopen{|}m\mathclose{|}=1 . The unstable frequency range of wind turbine wakes falls typically within 0.2<St<0.6 ( St=fD/U_{\infty} is the Strouhal number, with f the control frequency, D the rotor diameter, and U_{\infty} the oncoming wind speed) ( Cheung et al. 2024 ) . Numerical simulations and experiments have demonstrated that both strong helical meandering triggered by DIPC ( Frederik et al. 2020 ; Van der Hoek et al. 2024 ) and large side-to-side meandering triggered by rotor sway and yaw fall in this frequency range ( Li et al. 2022 ; Lin & Porté-Agel 2022 ; Messmer et al. 2024 ) . Despite its success, the eigenvalue nature of the LSA ( Schmid & Henningson 2001 ) limits its predictive capability: neither AWC-specific wake fluctuation nor enhanced wake recovery enhancement can be predicted by LSA. • Coherent velocity fluctuation: AWC-induced wake fluctuation is characterized by large-scale coherent structures ( Frederik et al. 2020 ; Li et al. 2022 ) . Unlike the sensitive frequency range, the coherent fluctuation is not only dependent on the wake’s instability property but also on specific rotor actuation ( Fontanella et al. 2021 ; Kopperstad et al. 2020 ; Fu et al. 2019 ) . For this reason, wake models for predicting coherent structures must function as an input-output system that links the rotor actuation and the wake fluctuating response. The resolvent analysis ( Trefethen et al. 1993 ; Sipp et al. 2010 ; McKeon & Sharma 2010 ) serves as an effective tool for this purpose. The resolvent analysis has been applied to explain the origin of coherent structures in wind turbine wakes generated by ambient turbulence ( De Cillis et al. 2022 ; Feng et al. 2022 ; Feng

Read the complete text