Dear Jason,
I have encountered two problems about FAST, and I think you can share your experience in this field
1.First, My model has only one single monopile, no tower and no wind turbine on top.In Stand-Alone SubDyn modules, I want to build a single monopile , and the water depth is 20 m, upper extend to 10 meters to TP, but, how can add wave load alone in Stand-Alone SubDyn modules, I want to examine the effect of wave load for the single monopile. If not achieve them in Stand-Alone SubDyn modules, So how can I achieve the effect of wave load on the single monopile in SubDyn Coupled to FAST.
2.There is a issue about in turbsim, The turbulence model has these options"IECKAI" TurbModel - Turbulence model (“IECKAI”=Kaimal, “IECVKM”=von Karman, “GP_LLJ”, “NWTCUP”, “SMOOTH”, “WF_UPW”, “WF_07D”, “WF_14D”, “TIDAL”, or “NONE”), but I want use Mann model this turbulence , which one should I choose, or what other way could I have my turbulence is Mann model .
Best regards,
Zhou Lin
Dear Zhou Lin,
Here are my answers to your questions:
-
The standalone SubDyn module does not consider waves or hydrodynanamic loads. Instead, you can set up a monopile without turbine in a coupled FAST model. To do so you’ll have to disable all turbine-related inputs e.g. rotor aerodynamics, controls (disable ServoDyn), blade/hub/nacelle mass/inertia, and blade/drivetrain/nacelle degrees-of-freedom.
-
The Mann model is not implemented in TurbSim, which is based on the Veers method (with prescribed settings of the wind profile + spectrum + coherence + Reynold’s stresses). However, you can use the Mann model in conjunction with FAST by using one of the freely available Mann turbulence simulators that DTU Wind Energy has made available, which output HAWC-formatted wind data files that can used directly by FAST’s InflowWind module.
Best regards,
Dear Jason,
Thank you for your quick reply, I understand the second question, but I have some doubts about the first one.
I want to set up a monopile without turbine, Whether the ElsotonDyn needs to be disable? instead,enable ElastoDyn ,disable these input files like blade/hub/nacelle mass/inertia, and blade/drivetrain/nacelle degrees-of-freedom. What should I do? Thank you for your advice.
Best regards,
Zhou Lin
Dear Zhou Lin,
Yes, you should enable ElastoDyn (actually it is not possible in the current version of FAST to disable ElastoDyn). You should enable all platform DOFs in ElastoDyn to ensure the top of the monpile is free to move (disabling the platform DOFs would constrain the top of the monopile). Most everything else in ElastoDyn can be set so that their effect is zero or negligible.
Best regards,
Dear Jason,
Thank for your reply, but,I have encountered some problems about my monopile model,
1.In your way, I have set up a monopile without turbine in a coupled FAST model. All of the mass and inertia are set to zero in my model, only platform 6 DOF set TRUE . but you known TipRad and HubRad aren’t zero, and TipRad > HubRad , and TowerHt can’t be set to zero, and TowerHt>TowerBsHt ,and TowerHt+Twr2shft+OverHang*SIN(shfttilt)>TipRad, so I set TowerHt=10.876,TipRad=0.63 HubRad=0.015, I run FAST after setting these parameters, and the operation can be successful, but the data will not converge. At t=0.1050s, the data will become NaN. My running results are shown in the following figure, and the files of my ElstoDyn module are as follows.
NRELOffshrBsline5MW_OC3Monopile_ElastoDyn.zip (3.24 KB)
2.Another question is whether the wall thickness of NREL5MW wind turbine Substructure is 27mm or 60mm?
Thank you very much,I wish you reply
Best regards,
Zhou LIn
Dear Zhou Lin,
My guess is your model is going numerically unstable due to an improper time step, but it is difficult for me to say more without more information.
Can you reduce further reduce the length of the geometries you are not modeling directly e.g. HubRad = 0, TipRad = 0.00001, TowerHt = 10.0001 m? Have you set the distributed masses of the tower and blades to near-zero? Is your PtfmYIner set correctly?
The thickness of the OC3-monopile for the NREL 5-MW turbine is 0.06 m.
Best regards,
Dear Jason,
Thank for you reply,I’ve done it your way, set HubRad = 0, TipRad = 0.001, TowerHt = 10.01 m,and set the distributed masses of the tower and blades to near-zero, and PtfmYIner set to the initial value 1.534E+06 of your source file.
I debugging repeatedly many times, I found .out file data is always not convergence, I also changed the different time time step, looks like the data or not convergence, I don’t know where the mistakes, or set unreasonable, the attachment is my model of the single pile complete file, I hope you can have a look, to know more information about my model, I wish you can solve my confused.thank you again!
Thank you very munch,I wish your reply.
Monopile.rar (2.84 MB)
Dear Zhou.Lin,
I tried to downloaded your archive, but the archive appears to be corrupted and I cannot access any files in it.
Best regards,
Dear Jason,
I am so sorry that my compressed file can not download,Attached is my new compressed file, I’m sorry.
If it is convenient, could you please tell me your work e-mail? If this file cannot be downloaded, I would like to contact you through the e-mail.
Thank you very much.
Main input files.zip (158 KB)
Best regards,
Zhou.Lin
Dear Zhuo.Lin,
I took a look at your model and got the model to run as expected with a few changes:
- In the FAST primary input file, set DT = 0.002, UJacSclFact = 3E5, and DT_Out = 0.002
- In ElastoDyn, set PtfmHeave = -1.19081E-4, PtfmRefzt = 10, and PtfmYawIner = 0
- In HydroDyn, set AddBLin(3,3) = 546230
This value of DT comes from the value recommended in section 5.8 of the draft SubDyn User’s Guide and Theory Manual: wind.nrel.gov/nwtc/docs/README_FAST8.pdf.
I hope that helps.
Best regards,
Dear Jason,
Thank you for you reply, my model works well, thank you very much for your help, but there is a problem.
I output the Fore-Aft Bending Moment at the MSL, and intercept the simulated time of 1200s, in which the time history curve of 760s-800s is amplified, as shown in the following figure, a problem is found.
You can see it in the picture, probably every 0.25 s curve will have a cycle of circulation, the corresponding spectrum is about 4 Hz, I guess there are two kinds reason. First, my model will exist in the process of running the high frequency motion, the cause of the high frequency motion I don’t know, if this is the case, hope you can put forward some Suggestions, thank you. Another guess is that I have a problem with the simulation time step,perhaps due to the instability of the numerical calculation resulting in this situation.
I hope you can give me some advice on this issue, thank you!
Best regards,
Zhou.Lin
Dear Zhou.Lin,
I don’t see any signs of a numerical instability in your solution. My guess is that you have a natural frequency in your FAST model that is around 4 Hz. You could add damping to the mode that is excited if you feel that the mode is overly excited.
Best regards,
Dear Jason,
Thank for your quick reply,
As you said, I’m trying to increase the damping to eliminate the excited, but the question is, should I add damping in the SubDyn module or should I add additional damping in the HydroDyn module?
If the damping is added on the monopile, could you tell me how to increase the corresponding damping on a monopile? I am sorry that I can’t find the damping term that can be increased in the SubDyn module.
Thank you for your suggestion. Your suggestion has been a great help to me. Thank you again.
Best regards,
Zhou.Lin
Dear Zhou.Lin,
I can’t tell you where to add the damping because I don’t know what mode is being excited your model. You’ll have to identify what mode is being excited first.
The structural damping of the Craig-Bampton (Internal) modes in SubDyn is set by input parameter JDampings in the SubDyn input file.
Best regards,
Dear Jason,
Thank for your quick reply, I appreciate your help very much. But I encounter a new problem.
as you known I’m currently researching on the substructure part of OC3-monopile (under SWL) where I need WAMIT input and output files to import hydrodynamic data into HydroDyn coupled to FAST. However, I do not have access to WAMIT, and in the HydroData file of the FAST v8 archive,I can’t find the WAMIT input and output file about OC3-monopile. Is it possible for NREL to provide me with these files(*.1, *.3, *.hst files and *.ms2, *.gdf files and etc. ) so that I could use the data for HydroDyn coupled to FAST.
Thanks a lot in advance. Thank you agian.
Best regards,
Zhou.Lin
Dear Zhou.Lin,
The HydroDyn hydrodynamics module of FAST supports a potential-flow solution (WAMIT interface), a strip-theory solution, or a hybrid combination of those two. For the OC3-monopile (and most any monopile), we model the hydrodynamics via the strip-theory solution, so, no WAMIT model is necessary.
Best regards,
