Airfoil characteristics for use in the Blade Element Momentum (BEM) method calculating the forces on Horizontal Axis Wind Turbines (HAWT) are derived by use of systematic methods. The investigation and derivation of the airfoil characteristics are basedon four different methods: 1) Inverse momentum theory, 2) Actuator disc theory, 3) Numerical optimisation and 4) Quasi-3D CFD computations. The two former methods are based on 3D CFD computations and wind tunnel measurements on a 41-m full-scale rotorwith LM 19.1 blades. The derived airfoil characteristics show that the lift coefficient in stall at the tip is low and that it is high at the root compared to 2D airfoil characteristics. The use of these characteristics in aeroelastic calculations shows agood agreement in power and flap moments with measurements. Furthermore, a fatigue analysis shows a reduction in the loads of up to 15 % compared to a commonly used set of airfoil characteristics. The numerical optimisation is based on both the 3D CFDcomputations and measurements on a 41-m rotor with LM 19.1 and LM 19.0 blades, respectively. The method requires power and loads from a turbine and is promising since a set of lift and drag curves is derived that can be used to calculate mean values ofpower and loads. The lift in stall at the tip is low and at the root it is high compared to 2D airfoil characteristics. In particular the power curves were well calculated by use of the optimised airfoil characteristics. In the quasi-3D CFDcomputations, the airfoil characteristics are derived directly. This Navier-Stokes model takes into account rotational and 3D effects. The model enables the study of the rotational effect of a rotor blade at computing costs similar to what is typical for2D airfoil calculations. The depicted results show that the model is capable of determining the correct qualitative behaviour for airfoils subject to rotation. The method shows that lift is high at the root compared to 2D airfoil characteristics. Thedifferent systematic methods show the importance of rotational and 3D effects on rotors. Furthermore, the methods show high lift coefficients in stall at the inboard part of the blade and low lift coefficients in stall at the outboard part of the bladecompared to 2D wind tunnel measurements.
Vindenergi og atmosfæriske processer; Risø-R-1065; Risø-R-1065(EN)