Issue with Wake Visualization in FAST.Farm: Alternating Results in OXY Plane

[Steven647]

Hello Everyone，

I'm a Ph.D. student studying the wake of floating wind turbine. I am passionate about using FAST.Farm for wake simulation, and I need some assistance with FAST.Farm. I am using FAST.Farm to simulate the wake of a wind turbine with harmonic motion， achieved through the ExtPtfm module in OpenFAST (as discussed in issue617 and this forum post)

The wake calculation domain set up and the wake dynamic properties in FAST.Farm follow the work published by Carmo and Jonkman in 2024. You can find the reference paper here:
Investigationg the interactions between wakes and floating wind turbines using FAST.Farm.pdf.

Problem Description:
I am simulating the wake of the NREL 5MW baseline wind turbine model, with the following settings:
InflowWind: WindType = 1, Wind speed = 11.4 m/s.
FAST.Farm Input File: Mod_Wake = 2 (Curled Model), dr = 12, NumRadii = 18, NumPlanes = 90, f_c = 0.01, C_Meander = 2.1.
The input file of FAST.Farm is as follows:
FAST.Farm.fstf.txt

FAST.Farm can successfully terminate the calculation, but the output visualization results are problematic. Specifically, Figure 1 and Figure 2 (below) alternate between steps in the OXY plane at z = hub height. This alternating pattern seems abnormal.

Output Visuallizations:
Fig. 1 The last step results of wake in OXY Plane（z=hub height）

Fig. 2 The next step results of wake in OXY Plane（z=hub height）

Question:
Can anyone help me understand why this issue occurs and suggest a solution? I would greatly appreciate any suggestions or insights！

Best regards,
Steven Liu

[jjonkman]

Dear @Steven647,

I agree that your model settings sound consistent with Wind Energy Science paper by Carmo et al 2024, other than the use of steady, uniform inflow. FYI: This paper is now published: https://wes.copernicus.org/articles/9/1827/2024/wes-9-1827-2024.html.

I'm not sure I understand what you are plotting. What do you mean by "last step" in Figure 1 and "next step" in Figure 2?

Can you also clarify what harmonic motion you are prescribing? Is this a prescribed harmonic surge motion of a given amplitude and frequency (if so what) or something else?

Best regards,

[Steven647]

Dear @jjonkman ,

Thank you very much for your reply. Please allow me to clarify the obtained visualization results of flow field. The wake results obtained based on the above settings are shown in the following animation:

This wake results seems abnormal. I am confused about what caused such wake results and unsure how to resolve it.

For the prescribed motion, a harmonic surge motion with an amplitude of 5 m and a frequency of 0.1 Hz is applied. The motion is achieved by defining an extremely large mass matrix in the ExtPtfm module, from which the stiffness and damping matrices are calculated, and then applying a harmonic load accordingly.

Best regards,

[jjonkman]

Dear @Steven647,

I'm not sure I understand your animation, e.g., why the domain size toggles between frames. It seems like you are grabbing planes of flow slices from different simulations or different cross planes and visualizing them together in one animation.

Thanks for clarifying your motion, but I note that your surge frequency is much higher than your low-pass cut-off frequency in FAST.Farm (f_c). Also, your low-resolution time step (DT_Low) will only sample motion surge motion 3.3333 times per cycle, so, the the wake calculation in FAST.Farm will be very much under-resolving your surge motion.

Best regards,

[Steven647]

Dear @jjonkman,

Thanks for your reply. As you mentioned, the strange wake animation occurred because I mixed calculation results from different simulations and visualized them using ParaView. I apologize for my lack of attention to detail.

For simulating the wake under harmonic motion using FAST.Farm, can I adjust the values of f_c and DT_Low based on my motion while still meeting the modeling guidance? For example, when prescribing a motion amplitude of 5 m and a frequency of 0.1 Hz, I set the low-pass cutoff frequency f_c to be slightly greater than 0.1 Hz and DT_Low to 2 s in order to capture the motion's influence on the wake. Are there any other parameter settings that require attention? Could you please provide some suggestion?

Best regards,

[jjonkman]

Dear @Steven647,

I'm glad your flow visualization now makes more sense to you.

You certainly can (and should) adopt DT_Low to match your own model requirements, with general modeling guidance documented here: https://openfast.readthedocs.io/en/main/source/user/fast.farm/ModelGuidance.html#low-resolution-domain.

I would be a bit careful setting f_c too high--especially when the rotor is undergoing large oscillatory motion--to avoid nonphysical wake effects, e.g., streaks showing up in the wake resulting from wake planes passing each other, as discussed in the above-mentioned paper.

Best regards,

[Steven647]

Dear @jjonkman,

Thank you very much for your professional advice. I will adjust the parameters and conduct the relevant simulations, and I will provide feedback once I obtain the results. Thanks again.

Best regards,