The aerodynamic part of HAWC2 is based on the principle blade element momentum theory (BEM). The classic approach is though modified to include wake expansion and swirl. Furthermore the BEM is extended with models to handle:

- Dynamic inflow

- Skew inflow

- Shear effect on induction

- Effect from large blade deflections

- Tip loss

- Three dynamic stall models:

- Stig Øye model

- Modified Beddoes-leishmann model

- ATEFlap dynamic stall model that includes the effects of active trailing edge flap deflection

**Dynamic inflow:
**

Dynamic inflow is a general term for the phenomena occurring when the load is changing on the rotor. If the changes are slow the induced velocity is equilibrium with the aerodynamic forces. But if the load changes are fast as e.g. a step change it will take some time before the induction correspond to the quasi steady value.

The classical blade element momentum theory was originally designed for axisymmetric flow. Often, however, wind turbines operate at yaw angles relative to the incoming wind, which produces a skewed wake behind the rotor. Therefore a correction is implemented to account for the skewed wake effect.

The BEM part of HAWC2 is based on calculation of thrust and corresponding induction in a polar grid. The thrust in a local grid point is obtained by calculating the thrust using the local grid wind speed, local grid induced velocity at the position of the two neighboring blades at same radial distance. Afterwards a linear interpolation of Ct on the two blades based on the azimuth distance between grid point and the two blades is used.

**Effect from large blade deflections
**Large flapwise blade deflections will cause a change in the effective rotor diameter, and blade forces will no longer be perpendicular to the rotor plane. This reduces the thrust on the rotor and is thereby changing the induced velocities and vice versa.

**Dynamic stall
**

Under normal conditions coefficients obtained from wind tunnel tests are sufficient to describe the characteristic of an airfoil. However when the angle of attack is changed rapidly a non-linear effect called dynamic stall occurs. This means that the stall phenomenon does not happen instantaneous but it takes some time for the flow to approach the stationary flow pattern. Two dynamic stall models have been implemented. The first one is generally known as the Øye model, which includes the effect of stall separation lag. The second model is a modified Beddoes-Leishmann model, which includes the effects of shed vorticity from the trailing edge (Theodorsen Theory), as well as the effects of stall separation lag caused by an instationary trailing edge separation point, influencing both lift, drag and pitch moment coefficients. These effects are especially important related to flutter analysis, but also generally to calculate loads and stability of blades with very low torsion stiffness. The ATEFlap dynamic stall model computes the unsteady lift, drag and moment on a 2D airfoil section equipped with Adaptive Trailing Edge Flap. The model captures the unsteady response related to the effects of the vor- ticity shed into the wake, and the dynamics of flow separation; a thin-airfoil potential flow model is merged with a dynamic

stall model of the Beddoes-Leishmann type.

The aerodynamic problem for vertical axis wind turbines is quite complex and different of a horizontal axis wind turbine. This is traditionally handled using double stream tube momentum modeling or using more complex aerodynamic methods like vortex and CFD simulation, whereas the latter are very computationally demanding.

The aerodynamic model for VAWT in HAWC2 is implemented by condensing the three dimensional VAWT to a number of 2D cylinders stacked vertically also called the 2D actuator cylinder model. For the 2D case it is possible to find the steady state solution in a fairly easy way using a modified linear solution of the actuator cylinder approach. The benefit of this is that it is a more physical correct solution than the double stream tube method and a very computational efficient and fast approach compared to vortex and full CFD solutions. The model is extended from a quasi-steady approach with a dynamic inflow model approach known from dynamic BEM formulations of HAWT simulations. Improvement of this model approach is still yet to be done, especially regarding finding the right time constants.