The actuator-disk method does not resolve the turbine geometry directly. Instead it uses tabulated lift and drag curves to determine the blade forces based on local flow conditions. In the simulation, these forces are imposed via. sources to the momentum equations in a sub-domain representing the rotor.
Two-equation turbulence models tend to over-predict the eddy-viscosity near the rotor (a region of high strain rate) and therefore over-predict the wake recovery rate. I'm testing some modified turbulence models that mitigate this problem.
I'm mostly interested in lab studies, but may consider field data as well. A good data-set needs:
1) Good characterization of the ambient flow, with measurements of the velocity profile (wrt. height from the floor/ground), tke and decay rate (which may be assessed by looking at tke at two or more streamwise locations).
2) Measurements of the turbine rotational speed and thrust force, ideally measuring the drag on supporting structures as well. Turbine power is also desirable (i.e. measured via torque and rotational speed).
3) Measurements of the wake lateral traverses over a range of downstream positions, ideally from 2 diameters (D) to 10D or further.
If you know of any published/unpublished or ongoing data campaigns that fit this description please let me know. I'm all setup to run simulations, and I'm interested in co-authoring papers.
Mr. Shives,
In the summer of 2013, I was a part of an internship at Suzlon Group (Pune, India) where we had developed statistical methods to account for shortfalls in turbine performance. We were granted access to extensive biannual of 10 turbines that included mean wind speed and direction at two heights, power output, nacelle angles, rotor speed, blade angle and a lot of other equipment-related parameters. I doubt if it includes wake-position readings.
As of now, I do not know if I have permission to share this data. If you are interested, I will talk to the concerned authorities and get back to you in a week's time.
Dear Michael,
Since more than 10 years NREL has been performing experiments to improve wind turbine designs and to provide validation data for classic and modern numerical models (including 3D CFD).
I suggest you to use their data to validate your model and later, use it with enough confidence to develop your own design under the obvious numerical uncertainties that any numerical method will have always attached (e.g., round-off, turbulence approximations and compressible treatment, for example).
NREL data of HAWT may be found at:
http://www.nrel.gov/docs/fy09osti/43508.pdf
Also, an example of validation of a CFD model using HAWT data can be found at:
https://ipfs.io/ipfs/QmfXH9XtP7xmoTH8WAp4HNSduqWMwLTH8B8TvbTkdgzNAa/cc-by-nc-sa-3.0/1404/1404.6183.pdf
and
http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.462.5540&rep=rep1&type=pdf
I hope it helps you.
Regards,
Luis
- Badges
- Science topic