Hello all,
I am trying to investigate the steady state equilibrium of wind turbine using spar buoy platform with catenary mooring cables.
To achieve this, I tried running FAST Test 24, by turning off CompInflow, CompAero, CompServo, CompSub, and CompIce, while setting WaveMod to 0 in HydroDyn.dat.
My understanding was that I am starting the computation with no external forces applied, with the condition of still water.
However, I am getting some oscillation in the Surge and Sway displacement of the platform, as can be seen in the attached figure.
Does this mean that FAST is starting the computation not in a static equilibrium state? Or are there some other implicit input parameters that I should turn off?
Thank you for your time and attention.
Best regards,
Yehezkiel Tumewu
M.S. Student Structural and Earthquake Engineering
Department of Civil, Structural and Environmental Engineering
University at Buffalo, The State University of New York
212 Ketter Hall, Buffalo, NY 14260
Dear Yehezkiel,
FAST is not initiated with a static solver; instead, the solution starts from the user-specified initial conditions. However, the OC3-Hywind spar model is close to static equilibrium. Its center of mass does not lie directly over the tower centerline, so, the system wants to natural pitch over (and surge) ever so slightly in the absence of environmental excitation.
I hope that helps.
Best regards,
Dear Jason,
Thank you for your reply and clarification.
I had a similar idea about the eccentricity of CoM and tower centerline, but I was not sure whether that was the cause, or not all the input parameters related to external loads are turned off.
I also did a simplified 2-D representation (3 DOFs: Surge, Heave, Pitch) of the NREL 5MW Tension Leg Platform (Test 23), to compare the natural frequencies with the ones generated from FAST. The resulting frequencies of the simplified model and FAST were quite similar.
However, this is not the case for the OC3-Hywind Spar (Test 24) and ITI Barge (Test 22).
Can you point me to some papers that discussed how FAST built the Stiffness and Mass matrices for the Spar-Buoy and Barge platforms?
Thank you for your time and kind attention.
I am looking forward to your reply.
Best,
Yehezkiel Tumewu
M.S. Student Structural and Earthquake Engineering
Department of Civil, Structural and Environmental Engineering
University at Buffalo, The State University of New York
Dear Yehezkiel,
I’m sorry, but I don’t really understand your question. What stiffness and mass matrices are you talking about?
Best regards,
Dear Jason,
My apologies for the confusion in the question.
I was referring to the mass and stiffness matrix of the wind turbines with spar-buoy platform and barge platform.
Attached to this post is the paper that I found which discussed the stiffness matrix of FOWT with Tension Leg Platform (TLP).
By using the stiffness and mass matrix, I obtained the natural frequency of the structure. The result was very similar with FAST’s resulting natural frequency.
However, for OC3-Hywind Spar (Test 24) and ITI Barge (Test 22) case, it wasn’t comparable.
I was wondering how FAST builds these matrices, or does it not use this method?
Thank you.
Best regards,
Yehezkiel Tumewu
M.S. Student Structural and Earthquake Engineering
Department of Civil, Structural and Environmental Engineering
University at Buffalo, The State University of New York
Linear Stiffness of TLP.pdf (515 KB)
Dear Yehezkiel,
I only skimmed the paper in your attachment, but FAST v8 does not form this global mass and stiffness matrix, at least not directly. In FAST v7, it was possible through a linearization analysis to export the equivalent linear global mass, stiffness, and damping matrices about an operating point, but this linearization functionality is not yet in FAST v8.
Best regards,
Dear Jason,
Thank you for taking the time to answer all my questions. It really clarifies some confusions that I had earlier.
May I ask another question related to your previous comment: “FAST is not initiated with a static solver; instead, the solution starts from the user-specified initial conditions”?
Can you point me where in the user guide that discusses this user-specified initial condition?
I am trying to start the computation at the equilibrium state (where the structure tilts by a small rotation due to its eccentricity from tower line), and afterwards compare the response.
Thank you very much.
Best regards,
Yehezkiel T.
Dear Yehezkiel,
The “Input Files” chapter of the old FAST User’s Guide describes each input parameter of FAST v7: wind.nrel.gov/nwtc/docs/README_FAST8.pdf. That said, there is nothing much to say about the initial conditions e.g. PtfmPitch defines the initial platform-pitch angle at time zero (the initial platform-pitch velocity is not entered in the input file, but is always assumed to be zero, unless the source code is changed accordingly).
Best regards,
Dear Jason,
Thank you for answering these barrage questions of mine.
It is more clear now after your explanations.
Dear Dr. Jonkman,
I would ask you if I change the mooring lines mass (varying both the length and the size) the static equilibrium position will change.
If so, if I wanted to set the system in the equilibrium I should add buoyancy or remove some ballasting. Am I right?
The question is how I can know this “weight difference” to set the correct static equilibrium position? The equilibrium cannot be calculated as a simply difference in weight, because the change in length of the cables will change the geometry, and so also the cable length resting on the seafloor (with its relative mass). Should I have to run an iterative calculation to look for the static equilibrium position for the varied mooring line system?
Please correct me if I making wrong reasonings.
Thank you very much.
Best regards,
Gabriele
Dear Gabriele,
The mooring system pretension (total vertical force acting on the platform from the mooring system) at the undisplaced position of the platform will be written to the MAP++ summary file. You can run MAP++ with both the original and redesigned mooring system to see how the total system pretension will change, from which you can decide how much mass to add/remove from the platform.
Best regards,
Dear Dr. Jonkman,
you mean run MAP++ standalone right?
Because if I run FAST with the mooring lines modified then the simulation crushed.
Is the Matlab routines in the MAP++ archive working? I have tried to compile MAP++, but I failed.
Do you a MAP++ executable or a working MATLAB script?
Thank you.
Best regards,
Gabriele
Dear Gabrielle,
I don’t have much experience with the standalone MAP++ executable. However, the MAP++ summary file is also written by FAST when running simulations with MAP++ enabled. The summary file is written at model initialization, so, as long as your simulation passes that step, you can get the information you need.
Best regards,
Dear Dr. Jonkman,
I written a MATLAB code that should read the MAP++ input file, modify it and then write a new .dat file to be run in FAST. The code paste all the lines excpets one, in which it modifies a numerical value (in this case the mooring lines length).
When I run FAST with this output file I obtain an error. Please see in attachment the screenshot and the MAP++ input file.
I have checked the input file and it seems exatcly the same.
Do you have any idea of why this happen?
Thank you.
Best regards,
Gabriele
output2.zip (528 Bytes)
Dear Gabriele,
It looks like your MAP++ input file is formatted incorrectly. When I open it up, there are no line breaks, meaning all data in the file is written to one line.
Best regards,
Hi all,
I have used the existed NREL 5MW mounted on OC3-Hywind spar buoy *.fts file from r-test repository. Although it runs with the initial time step of 0.0125s, I tried to change the timestep to 0.25s. I got the following error:
FAST_Solution:FAST_AdvanceStates:ED_ABM4:ED_AB4:ED_RK4:ED_CalcContStateDeriv:SetCoordSy:Small
angle assumption violated in SUBROUTINE SmllRotTrans() due to a large blade deflection (ElastoDyn
SetCoordSy). The solution may be inaccurate. Simulation continuing, but future warnings from
SmllRotTrans() will be suppressed.
Additional debugging message from SUBROUTINE SmllRotTrans(): 0.375 s
FAST_AdvanceStates:SolveOption2c_Inp2AD_SrvD:InflowWind_CalcOutput:CalcOutput:IfW_TSFFWind_CalcOut
put [position=(2.2556, -77.357, 106.19) in wind-file coordinates]: FF wind array boundaries
violated: Grid too small in Y direction. Y=-77.357; Y boundaries = [-72.5, 72.5]
FAST_AdvanceStates:AD_UpdateStates:BEMT_UpdateStates(node 5, blade 1):BEMT_UnCoupledSolve:There
is no valid value of phi for these operating conditions: Vx = -136.59, Vy = -882.38, rlocal =
11.547, theta = 0.25279, geometric phi = -2.988. This warning will not be repeated though the
condition may persist. (See GeomPhi output channel.)
FAST encountered an error at simulation time 0.25 of 5400 seconds.
Simulation error level: FATAL ERROR
Aborting OpenFAST.
Is there any explanation about it?
Thank you in advance.
Best regards,
Ioannis.
Hi Ioannis,
As discussed many times on this forum, the required time step is intrinsically tied to the frequencies inherent in the model. You can’t increase the time step without disabling functionality, e.g., structural degrees of freedom. If you simply increase the time step without disabling functionality, the solution can become numerically unstable.
Best regards,
Hello,
With respect which location are the platform DOFs (Surge, Sway, Heave, Roll, Pitch, Yaw) presented for OC3 Hywind Platform under forced vibration in OpenFAST? Is it the response at the COG or at the platform top?
Dear Akheel,
I’m not sure what you mean by “forced vibration”, as none of the OpenFAST r-tests have that. But the platform DOFs for the OC3-Hywind model in OpenFAST are defined about the point that is the intersection of the undisplaced tower centerline and mean sea level. This is because PtfmRefzt = 0 in ElastoDyn for this model.
Best regards,
Thank you sir
By forced vibration, I meant that the structure is subjected to external forces like wind and wave.
