Modelling of a 10MW FOWT

Discuss theory and modeling of wind-turbine structures.

Moderator: Bonnie.Jonkman

Yajun.Ren
Posts: 38
Joined: Mon Jan 21, 2019 6:54 am
Organization: The University of Edinburgh
Location: UK

Modelling of a 10MW FOWT

Postby Yajun.Ren » Mon May 18, 2020 10:31 am

Dear all,

I modelled a 10 MW wind turbine using the FAST input file given in https://rwt.windenergy.dtu.dk/natalia.c ... FAST_v1.00 and coupled it with a modified spar platform based on the OC3 hywind spar model. The hydrodynamic coefficients are obtained by WAMIT.

The results, especially the motions in 6 DOF seem very weird (please find attached files). The time series motions are more like free-decay shaped curve. In addition, the curves are not sine shaped, but square wave shaped.

I checked the input files and have no idea what I should do to correct the model, could you please give any suggests on this?

Regards,
Yajun
Attachments
10MW SPAR MOTIONS.zip
(55.17 KiB) Downloaded 70 times

Jason.Jonkman
Posts: 5771
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: Modelling of a 10MW FOWT

Postby Jason.Jonkman » Mon May 18, 2020 3:52 pm

Dear Yajun,

I've not used this FAST model myself, but are you not expecting free-decay behavior based on your simulation set up?

The "square wave shape" is likely the result of too little precision in the output file. Are you using a text file format (OutFileFmt = 1) with the default "ES10.3E2" OutFmt, which is limited to 4 digits of precision? You could increase this to, e.g., 6 digits of precision (changing OutFmt to "ES12.5E2"), or output in binary format (OutFileFmt = 2), which has full precision.

Best regards,
Jason Jonkman, Ph.D.
Senior Engineer | National Wind Technology Center (NWTC)

National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway | Golden, CO 80401
+1 (303) 384 – 7026 | Fax: +1 (303) 384 – 6901
nwtc.nrel.gov

Yajun.Ren
Posts: 38
Joined: Mon Jan 21, 2019 6:54 am
Organization: The University of Edinburgh
Location: UK

Re: Modelling of a 10MW FOWT

Postby Yajun.Ren » Tue May 19, 2020 3:31 am

Dear Jason,

Thanks for your reply.

No the structure was supposed to stay steady as I disabled AeroDyn, set WaveMod = 0 and the simulation started with the tubine at its undisplaced position. If it was a free decay, the phenomenon is not supposed to continue until the end of simulation (1000s). The initial heave motion is very large, which can be up to over -60 at the early stage of the simulation. Would you have any idea what can cause such a large heave motion? I tried to disabled the mooring system, the initial heave reduces about a half, but still shows the free-decay behavior.

According to your suggestion, I checked the output file in binary format and the curves look more reasonable.

Best regards,
Yajun

Yajun.Ren
Posts: 38
Joined: Mon Jan 21, 2019 6:54 am
Organization: The University of Edinburgh
Location: UK

Re: Modelling of a 10MW FOWT

Postby Yajun.Ren » Tue May 19, 2020 5:48 am

Dear Jason,

I would like to add two more questions:

I am not quite understand how you determine the value PtfmVol0 in HydroDyn. According to the manual, it is defined based on the result of WAMIT simulation. However when I checked the WAMIT output file published here https://rwt.windenergy.dtu.dk/natalia.c ... 0/Platform, it seems that the value of Volumes are not the same as the value of PtfmVol0 in the HydroDyn input file.

Secondly, I copied the relevant input files from https://rwt.windenergy.dtu.dk/dtu10mw/d ... FAST_v1.00 and try to reproduce the simulation of the DTU 10MW RWT with the NAUTILUS semi-submersible platform (static case), but the results also show a free decay behaviour (please find the attached file of surge and heave motions). This is hard to understand for me, would you suppose the results are reasonable? If not, what could the problem be?

Best regards,
Yajun
Attachments
heave_semi.jpg
heave_semi.jpg (45.34 KiB) Viewed 2095 times
surge_semi.jpg
surge_semi.jpg (40.97 KiB) Viewed 2095 times

Jason.Jonkman
Posts: 5771
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: Modelling of a 10MW FOWT

Postby Jason.Jonkman » Tue May 19, 2020 7:10 am

Dear Yajun,

Even in the absence of wind and wave excitation, i would expect a small amount of free-decay motion in every simulation, unless the initial conditions for all system displacements are set to their respective static-equilibrium values. For a floating wind turbine, there is often a small amount of surge and pitch free decay due to the overhanging weight of the rotor-nacelle assembly. I would not normally expect much initial free-decay behavior in heave if the floating system is set up as recommended, i.e., with the system weight balanced with the static buoyancy and vertical mooring pretension. In your results posted May 18, 2020, there is a large heave motion, which suggests the weight, static buoyancy, or moorings are not defined properly in the model. In your results posted May 19, 2020, there is a small heave motion, which suggests only a minor imbalance in the weight, buoyancy, and pretension.

I'm not familiar with the models you are using, but I would ask the original developer of that model for the reason in the discrepancy between PtfmVol0 and what is in the WAMIT output.

Best regards,
Jason Jonkman, Ph.D.
Senior Engineer | National Wind Technology Center (NWTC)

National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway | Golden, CO 80401
+1 (303) 384 – 7026 | Fax: +1 (303) 384 – 6901
nwtc.nrel.gov

Abhinay.Goga
Posts: 40
Joined: Tue Oct 25, 2016 8:12 am
Organization: SETEC elektronische Antriebsregelung GmbH
Location: Germany

Re: Modelling of a 10MW FOWT

Postby Abhinay.Goga » Thu May 21, 2020 4:34 am

Dear Yajun,

The free decay motions that you are experiencing could be because of the initial platform displacements given in the ElastoDyn. Try to verify static condition with all initial displacements zero. If the outputs of platform displacements are having non-zero values means the coupled structure is unbalanced due to buoyancy or some other reasons.

PtfmVol0 is the input for calculating initial center of buoyancy to balance the structure. Multiply that value with water density you will get displaced mass of water, which is supposed to be in-line with the coupled structure mass(Blades, RNA, Tower, Platform, Mooring lines). Could be the imbalance between these two that causes high heave.

Cheers

Ioannis.Bertsios
Posts: 11
Joined: Fri Mar 20, 2020 7:36 am
Organization: self
Location: Denmark

Re: Modelling of a 10MW FOWT

Postby Ioannis.Bertsios » Tue Aug 18, 2020 3:55 am

Dear all,

I am also trying to couple spar OC3-Hywind with DTU10MW RWT. During this try I got the following error. Could you please help to interpret it in order to be able to run this simulation.

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
OpenFAST

Copyright (C) National Renewable Energy Laboratory
Copyright (C) Envision Energy USA LTD

This program is licensed under Apache License Version 2.0 and comes with ABSOLUTELY NO WARRANTY.
See the "LICENSE" file distributed with this software for details.
**************************************************************************************************

OpenFAST-v2.3.0
Compile Info:
- Architecture: 32 bit
- Precision: single
- Date: Apr 03 2020
- Time: 14:10:27
Execution Info:
- Date: 08/18/2020
- Time: 11:31:25+0200

OpenFAST input file heading:
------- FAST model for DTU 10MW RWT mounted on spar-buoy from OC3-Hywind project ---------

Running ElastoDyn.
Running AeroDyn14.
Running InflowWind.

Reading a 33x33 grid (200 m wide, 9.99451E-03 m to 229 m above ground) with a characteristic
wind speed of 5 m/s. This full-field file was generated by TurbSim (v1.06.00, 21-Sep-2012) on
18-Aug-2020 at 11:07:00.

Processed 108800 time steps of 20-Hz full-field data (5440 seconds).
Running ServoDyn.
Running ServoDyn Interface for Bladed Controllers.
Running HydroDyn.
Generating incident wave kinematics and current time history.
Reading in WAMIT output with root name ".\../Platform/Spar".
Computing radiation impulse response functions and wave diffraction forces.
Running MoorDyn.
Creating mooring system. 3 fairleads, 3 anchors, 0 connects.
Finalizing ICs using dynamic relaxation.
t=19.001 FairTen 1: 9.10157E+05, 9.10661E+05, 9.10842E+05
Fairlead tensions converged to 0.1% after 19.001 seconds.
Basic DTU Wind Energy Controller (ver. 1.5) loaded...
PIDvar%outres= NaN
PIDvar%outpro= -1.41192486796464
PIDvar%outdif= NaN
dt= 0.000000000000000E+000
PIDvar%error1_old(1)= 0.000000000000000E+000
error(1)= -1.00500000000000
PIDvar%Kdif(1)= 0.000000000000000E+000
Padded_term= 0.000000000000000E+000
PIDvar%outset= 0.000000000000000E+000


FAST_Solution:CalcOutputs_And_SolveForInputs:SolveOption2:AD14_CalcOutput:AeroDyn was designed
for an explicit-loose coupling scheme. Using last calculated values from AeroDyn on all
subsequent calls until time is advanced. Warning will not be displayed again.
SolveOption2:SrvD_CalcOutput:BladedInterface option was designed for an explicit-loose coupling
scheme. Using last calculated values from DLL on all subsequent calls until time is advanced.
Warning will not be displayed again.

WARNING: High VNB velocity encountered during induction factor calculation.
Blade number 1, Element number 21
VNW = 4.553, VNB = -116.28
WARNING: High VNB velocity encountered during induction factor calculation.
Blade number 1, Element number 22
VNW = 4.3429, VNB = -131.54
WARNING: High VNB velocity encountered during induction factor calculation.
Blade number 1, Element number 23
VNW = 4.0304, VNB = -143.13
WARNING: High VNB velocity encountered during induction factor calculation.
Blade number 1, Element number 24
VNW = 3.7313, VNB = -151.71
WARNING: High VNB velocity encountered during induction factor calculation.
Blade number 1, Element number 25
VNW = 3.6599, VNB = -154.58
WARNING: Induced velocity warning written 5 times. The message will not be repeated, though
the condition may persist.

FAST_Solution:FAST_AdvanceStates:ED_ABM4:ED_CalcContStateDeriv:SetCoordSy:Small angle assumption
violated in SUBROUTINE SmllRotTrans() due to a large tower 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_Solution:CalcOutputs_And_SolveForInputs:SolveOption2:AD14_CalcOutput:ELEMFRC:BeddoesModel:ATT
ACH: Blade #1 element #20 is supersonic! Other elements are likely supersonic as well.
Supersonic mach nos. will be set to 0.7 to attempt continuation.
AD14_CalcOutput:ELEMFRC:BeddoesModel:ATTACH: Supersonic condition has subsided with Blade #1
element #33.
AD14_CalcOutput:ELEMFRC:BeddoesModel:ATTACH: Blade #2 element #23 is supersonic! Other elements
are likely supersonic as well. Supersonic mach nos. will be set to 0.7 to attempt continuation.
AD14_CalcOutput:ELEMFRC:BeddoesModel:ATTACH: Supersonic condition has subsided with Blade #3
element #4.
AD14_CalcOutput:ELEMFRC:BeddoesModel:ATTACH: Blade #3 element #20 is supersonic! Other elements
are likely supersonic as well. Supersonic mach nos. will be set to 0.7 to attempt continuation.
AD14_CalcOutput:ELEMFRC:BeddoesModel:SEPAR: Angle of attack = -6.15209E+10 is outside table.

FAST encountered an error at simulation time 0.375 of 5400 seconds.
Simulation error level: FATAL ERROR

Aborting OpenFAST.

--------------------------------------------------------------------------------------------------------------------------------------------------------------------------------

Thank you in advance.

Best regards,
Ioannis.

EDIT: Is there any possibility this issue to be related to wind speed grid?

Jason.Jonkman
Posts: 5771
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: Modelling of a 10MW FOWT

Postby Jason.Jonkman » Tue Aug 18, 2020 6:52 am

Dear Ioannis,

I doubt the problem is related to the turbulence. The model appears to be set up incorrectly and is either physically or numerically unstable. The first problem I see is that the controller is returning NaNs, which could point to a problem in the controller or model initial conditions. Does the simulation run if you disable the controller (by disabling the generator and nacelle-yaw degrees in ElastoDyn, GenDOF = YawDOF = False, and by disabling ServoDyn, CompServo = 0)? Could the time step by set too large; does reducing the time step allow the simulation to run?

Best regards,
Jason Jonkman, Ph.D.
Senior Engineer | National Wind Technology Center (NWTC)

National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway | Golden, CO 80401
+1 (303) 384 – 7026 | Fax: +1 (303) 384 – 6901
nwtc.nrel.gov

Ioannis.Bertsios
Posts: 11
Joined: Fri Mar 20, 2020 7:36 am
Organization: self
Location: Denmark

Re: Modelling of a 10MW FOWT

Postby Ioannis.Bertsios » Tue Aug 18, 2020 7:00 am

Dear Jason,

Thank you for your reply. I had an error in ElastoDyn input file regarding the DTU10MW RWT. I fixed it and I have received the following error

FAST_Solution:FAST_AdvanceStates:SolveOption2c_Inp2AD_SrvD:InflowWind_CalcOutput:CalcOutput:IfW_TS
FFWind_CalcOutput [position=(-14.703, -41.003, -0.33432) in wind-file coordinates]: FF wind array
boundaries violated. Grid too small in Z direction (height (Z=-0.33432 m) is below the grid and
no tower points are defined).

FAST encountered an error at simulation time 7.5 of 5400 seconds.
Simulation error level: FATAL ERROR

Aborting OpenFAST.

I have read in others posts that Z<0 means excessive deflection and I am trying to figure out how to solve this problem. If you have any idea how to fix it, please let me know.

Thank you in advance.

Best regards,
Ioannis.

Jason.Jonkman
Posts: 5771
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: Modelling of a 10MW FOWT

Postby Jason.Jonkman » Tue Aug 18, 2020 9:46 am

Dear Ioannis,

In addition to large deflection, this error could also be sensitive to whether you have tower aerodynamic analysis nodes in AeroDyn. Eliminating tower aerodynamic analysis nodes (or at least, eliminating the lowest-most nodes or moving them upward) may also solve the problem.

Best regards,
Jason Jonkman, Ph.D.
Senior Engineer | National Wind Technology Center (NWTC)

National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway | Golden, CO 80401
+1 (303) 384 – 7026 | Fax: +1 (303) 384 – 6901
nwtc.nrel.gov

Ioannis.Bertsios
Posts: 11
Joined: Fri Mar 20, 2020 7:36 am
Organization: self
Location: Denmark

Re: Modelling of a 10MW FOWT

Postby Ioannis.Bertsios » Wed Aug 19, 2020 12:09 am

Dear Jason,

Thank you for your reply. I will follow wat you said and hopefully I will solve this problem. Another thing that I observed is that OC3 Spar floater has a diameter of 6.5m in the connection zone with the wind turbine. NREL has a bottom tower diameter of 6m, but DTU10MW RWT has a bottom tower diameter of 8.3m. The reason why I would like to use the OC3-Hywind Spar floater is that I do not have access to WAMIT in order to calculate new matrices .hst, .1 and .3 output files.

Best regards,
Ioannis.

EDIT: I think you are right that the problem starts from AeroDyn module since I have received the following message:

FAST_Solution:CalcOutputs_And_SolveForInputs:SolveOption2:AD14_CalcOutput:AeroDyn was designed
for an explicit-loose coupling scheme. Using last calculated values from AeroDyn on all
subsequent calls until time is advanced. Warning will not be displayed again.
SolveOption2:SrvD_CalcOutput:BladedInterface option was designed for an explicit-loose coupling
scheme. Using last calculated values from DLL on all subsequent calls until time is advanced.
Warning will not be displayed again.
Time: 6 of 5400 seconds. Estimated final completion at 12:00:59.

Should I change the AeroDyn to v15 (I have checked that there is another post regarding this transition and the MATLAB-Toolbox)? Does AeroDyn v15 include unsteady aerodynamics?

Thank you in advance.

Jason.Jonkman
Posts: 5771
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: Modelling of a 10MW FOWT

Postby Jason.Jonkman » Wed Aug 19, 2020 7:08 am

Dear Ioannis,

You'll likely need to redesign the OC3-Hywind spar to support the DTU 10-MW reference wind turbine. Please note that it may be reasonable to model a spar buoy without potential flow (and thus not rely on WAMIT), relying only on the strip-theory solution. There are also publicly available alternatives to WAMIT, e.g., NEMOH or Capytaine (although these do not support the full capability of WAMIT).

You are receiving warnings about "explicit-loose coupling scheme" because your model includes corrections in the solver (NumCrctn > 0 in your OpenFAST input file). Changing from AeroDyn v14 to AeroDyn v15 will eliminate the first warning but not the second. You could also eliminate the corrections (NumCrctn = 0), which would likely require that you reduce the time step.

Yes, AeroDyn v15 supports unsteady airfoil aerodynamics, as well as dynamic inflow. I don't think switching from AeroDyn v14 to v15 will solve your immediate problem of an incorrect model set up, but we do recommend switching from AeroDyn v14 to AeroDyn v15 soon. We plan to deprecate AeroDyn v14 soon.

Best regards,
Jason Jonkman, Ph.D.
Senior Engineer | National Wind Technology Center (NWTC)

National Renewable Energy Laboratory (NREL)
15013 Denver West Parkway | Golden, CO 80401
+1 (303) 384 – 7026 | Fax: +1 (303) 384 – 6901
nwtc.nrel.gov


Return to “Structural Analysis”

Who is online

Users browsing this forum: No registered users and 1 guest