Hello,
I am a PhD student who is relatively new the field and to using OpenFAST. I am interested in modeling blade root moments (both flapwise and edgewise) for the IEA 15MW Monopile configuration. I have run a few simulations, and I am getting unexpected results, specifically for the edgewise root moment of the blades as the magnitude has significant oscillations in magnitude (but the blades are feathered so gravity effects should not be significant?). I am wondering if anyone has insight on why these oscillations are occurring or can point me in the direction of where I should check in the input files for errors. Here are some of the conditions and a figure I created that shows the root moments of each blade along with its position:
- Vhub = 25 m/s, Hs = 1.5
- Blade Pitch = 90 deg
- Idling
- ServoDyn is off
- Steady wind
*edgewise moment in blue
I am happy to share any other information or input files on the specific simulation.
Thanks!
Emma
Dear @Emma.Remien,
It appears the low-frequency variation in the moments (mostly flapwise due to the 90deg pitch) is coming from gravity, following the slow azimuth rotation you are showing. But I’m not sure why you are seeing a high-frequency oscillation.
Can you clarify more about your simulation set-up? Are you modeling the blade structural dynamics in BeamDyn or ElastoDyn? Do you see the same oscillation in other outputs such as the blade or tower deflections? What settings have you enabled in AeroDyn (for an idling rotor case, I would suggest Wake_Mod = 0 and UA_Mod = 0)?
I would also suggest computing a PSD of the signal to identify the frequency of excitation, which hopefully provide clarification on what is happening.
Best regards,
Hi @Jason.Jonkman ,
Thank you for your prompt response, I really appreciate it. The low frequency oscillation definitely makes sense.
I am modeling the blade structural dynamics in ElastoDyn. Additionally, in AeroDyn I already have Wake_Mod = 0 and UA_Mod = 0. There are some similar oscillations in other outputs such as blade deflection. Here are the outputs for the blade tip and towertop deflections:
I also computed the PSDs as seen below. I am not super well versed on this topic, however, so I may need some help understanding exactly what it means.
Thanks again!
Emma
Dear @Emma.Remien,
Thanks for clarifying.
The dominant frequency in the blade moment PSD appears to be at 1 Hz, which is odd to me because I’m not aware of a natural frequency at this frequency for the IEA Wind 15-MW RWT, e.g., as documented in the following paper: https://docs.nrel.gov/docs/fy24osti/88266.pdf. (Granted, the results in this paper are not for a pitch of 90degrees, so, maybe the frequencies shift a bit.) Presumably the oscillations go away if you disable the blade DOFs in ElastoDyn? Is there a specific DOF, when disabled, that eliminates the oscillations?
Best regards,
Dear @Jason.Jonkman ,
Thanks for the info. That does seem unusual.
I tried running the simulations with FlapDOF1 = False , FlapDOF2 = False, and EdgeDOF = False (separately), and they all seem to still have the high frequency oscillations. For the sake of runtime, I only simulated each for 300s instead of 800s, but you can still see the general trends.
FlapDOF1 = False
FlapDOF2 = False
EdgeDOF = False
Thanks!
Dear @Emma.Remien,
I’m still not sure what is happening. If you disable all blade flapwise and edgewise DOFs do you see still strong 1-Hz oscillations? How about if you disable the tower and platform DOFs? I would suggest trying various combinations of enabled/disabled DOFs to isolate the problem.
Best regards,
Dear @Jason.Jonkman ,
I was able to try a bunch of different combinations of DOFs on and off. The issue seems to be stemming from the PtfmHvDOF in ElastoDyn. When I turned that off and kept everything else on, the oscillations significantly reduced, and the 1-Hz oscillations are not present. Any insight on why this specific DOF would be having that effect?
Left (PtfmHvDOF = False), Right (PtfmHvDOF = True)
PSD with (PtfmHVDOF = False)
Thanks again!
Dear @Emma.Remien,
Thanks for tracking this down.
I would guess the issue is in the SubDyn Guyan mode associated with platform heave. Looking at the ElastoDyn input file provided for the IEA Wind 15-MW RWT atop the monopile (IEA-15-240-RWT/OpenFAST/IEA-15-240-RWT-Monopile/IEA-15-240-RWT-Monopile_ElastoDyn.dat at master · IEAWindSystems/IEA-15-240-RWT · GitHub), I don’t see that the SubDyn modeling guidance is followed for the setting of the initial platform heave displacement (PtfmHeave in ElastoDyn): 4.7.5. Modeling Considerations — OpenFAST v4.1.2 documentation. If you follow this guidance by setting PtfmHeave appropriately (when PtfmHvDOF is enabled), does the 1-Hz signal also go away?
Best regards,
