FLORIS wake effect simulation

[Deza.Nicolas]

Good afternoon,

I am analyzing the loss of yield of a wind farm due to the wake effect. To do this, I am carrying out simulations in FLORIS with parks formed by two turbines to study how the relative position of the upstream turbine decreases the performance of the downstream turbine. I am using the “Gauss-legacy” velocity model, the “Crespo Hernandez” turbulence model, the “Gauss” deflection model and the “sosfs” combination model.

In the attached image, the wind is blowing from left to right. The upstream turbine is located in the [0 , 0] position. And each square represents the power output of the downstream turbine, divided by the power output it would have without wake effect, depending on its position relative to the upstream turbine.

As can be seen in the image, the performance of the downstream turbine varies if its displacement (transversal to the wind) relative to the upstream turbine is -100m or +100m, or if it is -200m or +200m, etc.
Due to the transversal symmetry of the wake in the Gauss model, I understand that this should not happen with yaw angle equal to zero. Maybe it has something to do with the direction of rotation of the blades?

In any case, many thanks in advance.

Best regards,

Nicolás Deza

[Jason.Jonkman]

Paul Fleming of NREL has provided the following response:

This asymmetry is a result of the gauss-curl-hybrid (GCH) model which modifies a little bit the gaussian model, even when the upstream turbine is aligned. If you’d like to disable GCH set the following variables to False in the input JSON:

calculate_VW_velocities use_yaw_added_recovery use_secondary_steering

[Deza.Nicolas]

Thank you very much for the reply.

In such a case, to decide whether or not to deactivate GHC, would this asymmetry be observed in a real case with the wind turbine at Yaw = 0º? Or is it a model failure?