Swept Away

Abstract

In the transition from fossil fuels to renewable energy sources, the advancement and deployment of wind turbine technology plays a crucial role. A notable trend in wind turbine design is the ever-growing rotor size, which entails that wind turbine blades have become very slender and flexible structures.

This increasing flexibility offers an opportunity for tailoring the aeroelastic behaviour of wind turbine blades. One such aeroelastic tailoring technique is blade sweep, defined as a displacement of the blade axis in the rotor plane. Blade sweep couples bending and torsion deformations and can thus be used to passively alleviate loads on the blade.

For the design, optimisation, and certification of wind turbine blades, blade element momentum theory (BEM) remains the aerodynamic simulation method most relied upon. However, BEM-based numerical tools inherently assume a straight blade geometry and, hence, cannot accurately model the additional flow complexities introduced by blade sweep.

This dissertation starts by presenting a newly developed BEM correction model for swept blades. The focus is on accurately modelling the azimuthal displacement of trailed vorticity and the curved bound vortex self-induction while maintaining BEM's streamtube-independent approach and rapid calculation speed. The developed model shows good agreement with mid-fidelity modelling (lifting line simulations), which intrinsically can model the two aforementioned effects of blade sweep.

To validate the BEM correction model for swept blades beyond the comparison with lifting line simulations, two wind tunnel campaigns are conducted, one with straight blades, being thrust-scaled versions of the IEA 15 MW reference wind turbine blades, and one with swept blades. While the former is intended to provide a baseline for the accuracy of numerical modelling, the latter then provides means to assess the impact of blade sweep and how this is captured in low-fidelity numerical simulations. The validation is conducted based on blade-aerodynamic quantities derived from flow fields measured using particle image velocimetry (PIV). It is demonstrated that the application of the BEM correction model improves the match with the experimental data compared to simulations without the correction model being applied.

Furthermore, this dissertation covers three diverse research efforts conducted within the framework of the TIADE project, a field experiment on a full-scale wind turbine. Firstly, a robust approach to optimise the spacing of pressure sensors for aerodynamic measurements on wind turbine airfoils is presented. The approach considers the expected turbine operating conditions and improves the lift prediction accuracy compared to a simpler, cosine sensor spacing over a wide range of angles of attack. Given that two fundamentally different optimisation routines arrive at close-to-identical solutions, it can be concluded that an optimal solution exists for placing pressure sensors around an airfoil to conduct aerodynamic measurements.

Secondly, pressure measurements obtained on the TIADE research wind turbine over multiple months are employed to validate aeroelastic simulations. The validation is performed based on both ten-minute average data and time-resolved data and using both the integrated sectional forces and the underlying pressure distributions. Generally, a reasonably good agreement between simulated and measured data is found. This indicates that BEM-based aeroelastic algorithms are still valid tools to simulate modern, multi-megawatt wind turbines and their slender and flexible blades.

Finally, a design study of a blade with swept tip for the TIADE field experiment and thus under realistic geometric and load restrictions is conducted. Simulations suggest that flapwise fatigue and extreme blade root loads can be reduced. The same holds for the fore-aft and yawing moments at the turbine tower base. Simultaneously, the turbine performance in terms of power output remains unaffected. These results highlight the potential benefits of blade sweep as an alternative tip geometry for modular blades or as a conscious design choice for future generations of blades.

In conclusion, this dissertation contributes to a more accurate understanding and numerical modelling of swept blade aerodynamics. By moving from fundamental analyses all the way to more applied investigations of swept blade tips for a field experiment, the presented research helps pave the way towards swept blades being a valid option in future wind turbine designs.