Wind Energy (Apr 2021)

Simulation of oscillating trailing edge flaps on wind turbine blades using ranging fidelity tools

  • John M. Prospathopoulos,
  • Vasilis A. Riziotis,
  • Eva Schwarz,
  • Thanasis Barlas,
  • Maria Aparicio‐Sanchez,
  • George Papadakis,
  • Dimitris Manolas,
  • Georg Pirrung,
  • Thorsten Lutz

DOI
https://doi.org/10.1002/we.2578
Journal volume & issue
Vol. 24, no. 4
pp. 357 – 378

Abstract

Read online

Abstract Recent research developments have indicated that substantial reduction of both the fatigue and ultimate loads can be achieved by adopting trailing edge (TE) flap control strategies. Their aeroelastic tools employ blade element momentum (BEM) aerodynamic models enhanced with a sectional 2D treatment of the TE flap, neglecting the 3D effect of the trailed vorticity in the vicinity of the moving flap. In the present paper, a cross comparison of the BEM‐based models used in the aeroelastic analysis tools against higher fidelity, free‐wake lifting line, and fully resolved CFD models is performed, with the aim to highlight limitations of the first. A second level of comparison assesses the differences among tools of the same level of fidelity from different research groups. Moreover, a number of engineering‐based correction models that are used in conjunction with BEM and account for the complex 3D trailed vorticity effect are assessed. Simulations of a stiff rotor configuration of the DTU 10 MW Reference Wind Turbine are performed for a prescribed, harmonic TE flap motion, and aerodynamic loads are compared at the sectional and rotor‐integrated level. For the studied stiff rotor with the chosen flaps configuration, the results of the code‐to‐code comparisons indicate that low‐fidelity BEM tools consistently predict 1P variations of the rotor thrust due to the TE flap motion, but fail to reproduce the details of the load distributions especially in the vicinity of the flap section. BEM‐based corrected models, which account for 3D‐induced velocity effects, provide load distribution predictions closer to higher fidelity free‐wake and CFD models.

Keywords