Dear all,
I tried to run the simulation for monopile IEA 22MW, I got the files from IEA-22-280-RWT/OpenFAST/IEA-22-280-RWT-Monopile at main · IEAWindTask37/IEA-22-280-RWT · GitHub and I use Openfast3.5.3 and Rosco 2.8.
I tried several attempts.
1- I enabled ROSCO 2.8, and used normal turbulence model with mean wind speed at hub height equal to rated wind speed 11 m/s. I changed the water depth in the IEA-22-280-RWT-Monopile.fst from 300 to 34 ( I believe that this value was for floating) and in ElastoDyn I changed the BlPitch=2.42859 and RotSpeed=6.7709 according to the turbine responce in the steady state table from the manual. I changed PtfmHeave according to the equation in the manual
I got this error , It couldn’t run and stop from time 0
2- I tried to simplify the model, I used steady wind instead of turbulent. I got this error
I don’t know why there is NAN! and the GenPwr and GenTq= zero! ( I didn’t change the DOF in ElastoDyn I keep GenDOF= TRUE)
3- I simplified the model more, I turned of the controller in the ServoDyn, PCMode=VSContrl=zero, and steady wind.
It terminated normally after 10 hours for 630 sec. (I red in the forum that it takes a long time)
But I got the GenPwr = negative!!
I don’t know why I have these errors.
Thanks in advance. I appreciate your help.
Marwa
Dear @Marwa.Mohamed,
I am a user of OpenFAST, not from NREL staff.
My first question is: Could you please tell us what you are trying to do with OpenFAST?
Second question, could you please explain what Ptfm Heave is used for ?
Third question, could you give more info about the sea state ?
Fourth question, could you share your .fst file ?
Best Regards,
Riad
Thanks for your reply @Riad.Elhamoud.
1- I want to study the load analysis of the turbine under different design load cases according to IEC 61400-3 and determine the critical load at the seabed to design the monopile.
2- in the model consideration of SunDyn they pointed to that
3- I used the metoceon data from IEA 15 MW manual, which I watch in webinar for IEA 22 MW that they used the same metoceon data
4- this is my glue code.fst
Regards,
Marwa
Dear @Marwa.Mohamed,
I download the IEA 22 MW OpenFAST model. I tried the model in a free decay where the TTDspFA at time equals zero in ElastoDyn file is set to 1 m in the absence of wind and wave.
-
In the .fst file, i disabled ServoDyn, AeroDyn, Inflow.
-
In the HydroDyn file, i set WaveMod to zero.
-
In the .fst file, i changed the time step and set DT to 0.01 sec.
-
In the .fst file, i set CompElast to 1. I realize through reading old posts in the forum that BeamDyn would slow the simulation so i disabled it.
The simulation duration is 630 sec.
As a result, the simulation is too slow compared with the 5 MW !!!
Maybe you could disable BeamDyn to speed up the simulation. In the ElastoDyn file, disable Drivetrain, teether, yaw of the nacelle and the generator as suggested by @Jason.Jonkman
Sorry i did not give you a straight answer. Hope that what i have said may help.
Best Regards,
Riad
Thanks Riad for your help and time
Dear @Marwa.Mohamed,
Your simulation results suggest that you’ve disabled the ROSCO controller, e.g., either by disabling ServoDyn altogether or by selecting a different pitch or torque control option in ServoDyn. If GenDOF = FALSE in ElastoDyn, the controller selection won’t have any effect on the simulation results (other than ServoDyn GenTq and GenPwr, which will be calculated, but not impact the generator speed), but when GenDOF = TRUE, the controller will strongly impact the system response.
Best regards,
Thanks Dr. Jason for your quick response and comment. I understand what you say, that’s why I 'm confused about the results. I enabled GenDOF = TRUE, and used ROSCO and I put the DLL_InFilein the correct path, OpenFast doesn’t complain that it couldn’t run ROSCO! I have attached my ElastoDyn and ServoDyn. I will try another time with testing NTM and see the results.
Reagards,
Marwa
Dear @Marwa.Mohamed,
Do you have ServoDyn enabled in the OpenFAST primary (.fst) input file (CompServo =1)?
Best regards,
yes
Thanks for your response!
Dear @Marwa.Mohamed,
Do the messages written to the command prompt when OpenFAST is executed mention that ROSCO has been called? Perhaps your are mixing up which ServoDyn input file OpenFAST is calling?
Best regards,
Dear Dr. Jason,
Thanks for your reply.
yes the command prompt mention that ROSCO has been called.
I tried yesterday one thing (which I know is wrong). I download the original files of IEA 22MW and I changes the water depth to 34 (keep the original initial conditions for Blade pitch angle and rotor speed as it is and the original steady wind 5m/s and I didn’t change any thing in the Degree of Freedom in ElastoDyn) BUT (in ServoDyn I used the IEA-15-240-RWT-Monopile_DISCON.IN and Cp_Ct_Cq.IEA15MW.txt ) from 15MW to see if again there will be zero generator power??
The model run for 10 hours as before but in this time there is power from the generator
another thing that in the command prompt the screen was as usual
generator speed…pitch angle…power…wind speed
but when using the correct files for 22MW
IEA-22-280-RWT_DISCON.IN, I
EA-22-280-RWT_Cp_Ct_Cq.txt the command prompt is strange
I don’t know why I have this error.
I will download the original files again and try for the last time and come back to you.
Regards,
Marwa
Dear Dr. Jason,
I download the files of IEA 22 MW again and run the analysis for 60 sec and water depth of 34 m, I leave all things as it is and used ROSCO 2.8.
AGAIN THE GENERATOR POWER= ZERO
Regards,
Marwa
Hello Marwa,
I’m quite puzzled by your errors. A few ideas to debug:
- Try running the OpenFAST model from the 22MW repo as is. It should generate power
- You don’t need to debug ServoDyn with BeamDyn turned on. Switch back CompElast to 1 and increase DT to 0.01s and run your tests quickly. Also, you can switch TMax to 10s or so and do quick checks. Power should not be 0 from the start
- For some colleagues I recently uploaded input and output files on WeTransfer. The link expires on Monday Unique Download Link | WeTransfer. The load setup is described in Radware Bot Manager Captcha,
section 3.5
Let us know when you get power (or if you still can’t)
Regards,
Pietro
Dear Dr. Pietro.Bortolotti,
Thanks for your response.
I tried your second solution used CompElast = 1, DT = 0.01s, TMax=60 sec. and there is no power
I don’t know why the command prompt looks strange when running the turbine with IEA-22-280-RWT_DISCON.IN , IEA-22-280-RWT_Cp_Ct_Cq.txt. or that’s Ok?
I tried your #3 recommendation, first I tried to understand the files and I found that each wind speed has its own files . So, I tried 4 cases iea-22_43, iea-22_75, iea-22_10, iea-22_00. I found that the 4 cases has the same files for (DISCON.IN and Cp_Ct_Cq.txt) the difference just in the naming. Than I compare these files by the original files (IEA-22-280-RWT_DISCON.IN and IEA-22-280-RWT_Cp_Ct_Cq.txt) which have been uploaded in Github for public. I found that IEA-22-280-RWT_DISCON.IN has many differences that’s why it has no generator power out put.
Then, I tried to run the files from IEA-22-280-RWT/OpenFAST/IEA-22-280-RWT-Monopile at main · IEAWindTask37/IEA-22-280-RWT · GitHub BUT this time I used DISCON.IN and Cp_Ct_Cq.txt from the files you shared in WeTransfer and I gut GenPwr FINALLY!!!
.
To duple check I went back to the files you shared by WeTransfer and tried to run one case which is iea-22_75 with the (IEA-22-280-RWT_DISCON.IN and IEA-22-280-RWT_Cp_Ct_Cq.txt) which have been uploaded in Github for public.
I used CompElast = 1, DT = 0.01s, TMax=60 sec., in iea-22_75_InflowFile.dat i used WindType=1. and there is no powerI couldn’t have GenPwr!!
and again the strange comand prompt
In conclusion, I think that the DISCON.IN that has been uploaded in Github has problem and when using I couldn’t have GenPwr and has strange comand prompt, but when usin the DISCON.IN that you shared in WeTransfer the turbine run normally!
Sorry that I talk a lot. I hope that I didn’t make you confused more.
Regards,
Marwa
Hello Marwa,
I’m glad you’ve found a way to solve the problem. I believe the issue is that you’ve been running an outdated version of ROSCO. You wrote that you are running ROSCO v2.8.0, but both DISCON files are meant for ROSCO v2.9.0 (see line 2 at the top of the files). I’ve run a quick check and I get power for both DISCONs (the one I uploaded to wetransfer and the one on GitHub) when I run ROSCO v2.9.0, but power in only the DISCON from wetransfer when I use ROSCO v2.8.0
Could you please confirm my hypothesis?
Best regards,
Pietro
1 Like
Dear Dr. Pietro.Bortolotti,
Thanks for your reply. Your hypothesis is right. when I tried ROSC 2.9.0. I got power and the command prompt run normally with DISCON.IN files on Github.
another question please, is there any way to speed up the simulation of IEA 22MW, Can I turn off BeamDyn or the results will be inaccurate due to the blade torsion.
Thanks for your consideration and help.
Regards,
Marwa
glad that you solve your problem !!!
Thanks for your help!
Regards,
Marwa
