Forum
Important Notice for New User Registrations
To combat an increasing number of spam and bot registrations, we now manually approve all new user registrations. While this may cause a delay until your account is approved, this step is essential to ensure the quality and security of this forum.
To help us verify your registration as legitimate, please use a clear name as user name or an official email address (such as a work, university, or similar address). If you’re concerned that we may not recognize your registration as non-spam, feel free to email us at with a request to approve your username.
The relationship between wake length, rotations and wind speed
Hi David
I want to calculate the wake field of 5MW wind turbine 10D, wind speed 11.4m/s, rotor speed 12.1 rpm, 6° calculation step, I think about 80 rots, that is, 4800 steps, I opened the vortex roll up and vortex stretching, including the shed vortex and trailing vortex, wake zone is divided into four zones set as 20 rot, the zone factor is 1.
This multi-threaded calculation has lasted more than ten hours, only half of the calculation.
I wonder if I’m a little wasteful in setting this up. Do you have any better recommendations? Another problem is how to know the relationship between the number of rotations and the length of the wake field calculation, for example, how many rotations do I need to calculate 10D?
BR,
Roby
Hi Roby,
4800 time steps doesnt sound like an excessive number for wake studies. Whats critical with respect to the computational cost is the total number of free vortex filaments in your simulation. To reduce the cost (without reducing wake length) you can make the near wake zone or wake zone 1 shorter, and increase the length of zones 3 & 4. Also, you can try reducing the number of blade panels – this also has direct influence on the total vortex filament number.
Whether 20 rotor revolutions are sufficient for a wake length of 10D (if extracting results at 10D the total wake should be longer than that) depends on the tip speed ratio at which the turbine is operating at. I would suggest creating a low fidelity turbine setup (maybe just 4-5 blade panels, no structural model) for testing how many revolutions you need for 10D.
BR,
David
Dear David
The first figure is the result of 2400 steps or so, the turbulence viscosity coefficient is 1000, there are some singularities in the wake, I wonder if it is because the induced velocity calculation is overdone.This is the result of opening the vortex stretch and the vortex roll up.
In the second figure, the vortex roll up and vortex stretch are not considered, and the whole wake field is only evolved as a vortex lattice.The wake is uniform, and the resulting velocity profile is also uniform.
In the velocity cup plan output , I want to know what the meaning of “all next” after “At time” is, refers to the output of the current number of seconds plus how many percent of the number of seconds, each second under the velocity plane? What does the latter average mean?
In addition, I chose to set a point for both x and z, and set many points for y, which is equivalent to detecting the velocity distribution of a line based on y coordinates behind the hub x. Is it reasonable that the wake field obtained directly in this way is not uniform?
Uploaded files:- You need to login to have access to uploads.
Hi,
if you sample the wake velocity from a single timestep only some “jumpiness” in the data can be expected. This is because the current position (or phase) of the helical vortex structure (the wake) will influence your velocity samples.
To get “smooth” velocity distributions you should average the velocity samples. To do this you need to store the replay of the simulation by activating the “Store Replay” option in the simulation settings. If you have the wake positions stored for all timesteps you can extract velocity data at each timestep and also do an automated averaging.
If you choose the “all next” option in the cut-plane dialog cut planes will automatically generated for all timesteps (starting from the timestep that is currently selected by the Replay slider at the top).
The “% x” option allows to automatically only evaluate cut planes for every xth timestep (cut planes will only be generated if mod(timestep,x) == 0).
The “average” option will automatically generate a cut-planes with velocities that are averaged over all cut-planes currently aveluated with the “all next” option. Furthermore, this option also calculates cut-planes that contain the turbulent fluctuations in the wake.
Did you know that you can also export and import cut-plane definitions? This allows you to somewhat automate you simulation post-processing in the GUI (or fully automate post-processing in the CLI).
Check this link for the documentation: https://docs.qblade.org/src/user/simulation/cutplanes.html
BR,
David
Hi David
Thanks for your answer, this feature is so cool.
BR,
Roby
