Dear Professor Jonkman,
Hi I hope you’re doing well.
Recently, I’ve been experimenting with a passive Tuned Mass Damper (TMD) applied to the NREL reference turbine, using parameters from a published paper. I tested the setup in both OpenFAST and FAST-SC, but surprisingly, I encountered similar issues in both cases.
Instead of reducing structural responses, the TMD seems to increase the tower top displacement, particularly in the fore-aft direction. To quantify the effect, I used RMS (Root Mean Square) to calculate the percentage reduction compared to the baseline, but the presence of the TMD consistently made things worse in both environments.
This left me with a few questions:
Could I be making mistakes in defining the TMD parameters or boundary conditions?
Is it possible that I’m misinterpreting the outputs in my calculations?
Are there important modeling considerations or common pitfalls when using the passive TMD models in OpenFAST or FAST-SC?
Any advice or references you could share would be a huge help. I’m really trying to understand what’s going wrong and would appreciate your expert opinion.
Thank you so much for your time and support
Best regards
Anise Ziamehr
Dear @Anis.Ziamehr,
It is hard for me to say without knowing more about your simulation set up. If you believe your model is set-up correctly, I would first verify that the response is converged, e.g., by dropping DT or adding a correction step (NumCrctn > 0). If the response is converged for different solver settings, I would double check the TMD settings.
Best regards,
Dear Professor Jonkman,
Hello again! I really appreciate your time and previous feedback.
To help clarify the issue I mentioned earlier regarding the passive TMD not improving the tower response, I’ve attached a screenshot showing the set of parameters I used in the simulation. These are the values I applied for the TMD configuration in both OpenFAST and FAST-SC.
Despite trying different tuning values, the TMD still causes an increase in the fore-aft tower-top displacement, and I’m really trying to understand what might be going wrong.
If you have a moment to glance at the attached data and let me know if something seems off or unusual, I’d be incredibly grateful.
Thank you so much again!
Best regards,
Anise Ziamehr
(attachments)
Dear @Anis.Ziamehr,
Looks like a simple set up with only a linear mass-spring damper. Presumably this is for a nacelle-based TMD (NumNStC = 1 in ServoDyn)?
Can you share the time-series response of tower-top displacement with and without the TMD enabled? Did you ensure that the solution is converged with lower DT and increased NumCrctn?
Best regards,
1 Like
Dear @Anis.Ziamehr,
Hope you are doing well.
Could you please share your ServoDyn input file ?
Before tuning a TMD, did yo run a dynamic analysis in spectral domain in order to know (i) what is the excited modes and (ii) what is the source of excitation of the FOWT ?
Best Regards,
Riad
Dear professor jonkman
Off course , I share you the time _series response of tower displacment Windspeed:12m/s
Tmax:300s
DT:0.0125
WaveHs:6m
WaveTp:10s NumCrctn:0 Tmd mass:28ton
Stiffness:400000
Damping:4000
X and Y:true
(Attachment (without control)5MW_TLP_DLL_WTurb_WavesIrr_WavesMulti (2).out is missing)
(Attachment (without control)5MW_TLP_DLL_WTurb_WavesIrr_WavesMulti.out is missing)
(Attachment compare with and without tmd.xlsx is missing)
Dear jonkmam
I forgot to sent that one in previous email
(Attachment (with control)5MW_TLP_DLL_WTurb_WavesIrr_WavesMulti.out is missing)
Dear professor jonkman
Off course , I share you the time _series response of tower displacement Windspeed:12m/s
Tmax:300s
DT:0.0125
WaveHs:6m
WaveTp:10s Tmd mass:28ton
Stiffness:400000
Damping:4000
X and Y:true
NumCrctn:0 actually you said I change this parameter but I don’t know what it could be the best for it.
as I cant attach an excel file here I sure an link of time series and results ,if it’s possible for you please check it.
Dear @Anis.Ziamehr,
It looks like your TMD is reducing tower side-side motion, as well as platform-roll:
Regarding NumCrctn, you could try NumCrctn = 1 and verify that the results are unchanged. If they do change, you could increase NumCrctn incrementally and try again.
Best regards,
