load simulation

Discuss theory and modeling of wind-turbine structures.

Moderator: Bonnie.Jonkman

Ning.Wang
Posts: 17
Joined: Tue Jul 30, 2019 12:52 am
Organization: Ecust
Location: china

load simulation

Postby Ning.Wang » Wed Oct 16, 2019 3:21 am

Hi,everyone:
When I was simulating with fast, there was a relatively large peak of about 1.5hz when looking at the rootmxb1 load. I couldn't find out which specific component was produced in the view of various files. It was preliminarily certain that the coupling between blade 1st edgewise and other working conditions, but I couldn't find the proper coupling between blade 1st and what coupling, nor was not sure whether this guess was correct. And using the fast model of the official website, it is shown that the peak of 1.5hz also exists, and it is independent of the wind speed and wind type.
Attachment is my simulation picture, because the forum doesn't seem to put the picture, please help me, thank you very much!

Jason.Jonkman
Posts: 4952
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: load simulation

Postby Jason.Jonkman » Wed Oct 16, 2019 6:37 am

Dear Ning,

I can't see your picture (neither in this forum post, nor in the related e-mail you sent me) and you haven't stated which wind turbine you are simulating. But I would guess the 1.5 Hz is either a natural frequency of the system or related to excitation frequency, i.e., at n*RotSPeed, where n is an integer multiple (n= 1 or 2 or 3...).

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

Ning.Wang
Posts: 17
Joined: Tue Jul 30, 2019 12:52 am
Organization: Ecust
Location: china

Re: load simulation

Postby Ning.Wang » Fri Oct 18, 2019 12:57 am

Hi,Dear Jason:
I'm so sorry that the pictures didn't show .Annex 1 is the result of recompilation after the speed increase of 10%,1.5Hz still has a peak.Annex 2 is the simulation results formed after changing the transmission chain,the degree of freedom of the tower,and Annex 3 is the simulation result of the official website model under steady wind.It is now possible to determine the type of wind the type of wind,the speed of wind,1P.
Thank you very much!
Attachments
2.png
2.png (158.16 KiB) Viewed 2142 times
1.PNG
1.PNG (106.62 KiB) Viewed 2142 times

Jason.Jonkman
Posts: 4952
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: load simulation

Postby Jason.Jonkman » Fri Oct 18, 2019 9:06 am

Dear Ning.Wang,

I guess I'm still not sure what question you are asking. Which wind turbine are you simulating and do you know its natural frequencies (at various rotor speeds)? What rotor speed (or speeds) are you simulating? Do the peaks in the PSD correspond to either of these frequencies, or integer multiples of the rotor speed?

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

Ning.Wang
Posts: 17
Joined: Tue Jul 30, 2019 12:52 am
Organization: Ecust
Location: china

Re: load simulation

Postby Ning.Wang » Sat Oct 19, 2019 10:07 pm

Hi,Dear Jason:
Attached is that I have summarized my ideas and problems, which may be clearer. Thank you very much for taking a look at it for me. Thank you very much!
Attachments
Problem combing.doc
(311 KiB) Downloaded 51 times

Jason.Jonkman
Posts: 4952
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: load simulation

Postby Jason.Jonkman » Sun Oct 20, 2019 5:33 pm

Dear Ning.Wang,

I would guess the approx. 1.5-Hz mode is the drivetrain torsion mode, which has little influence from rotational speed. To confirm this, I suggest running a separate simulation with the drivetrain-torsion degree of freedom (DOF) disabled (DrTrDOF = False).

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

Ning.Wang
Posts: 17
Joined: Tue Jul 30, 2019 12:52 am
Organization: Ecust
Location: china

Re: load simulation

Postby Ning.Wang » Sun Oct 20, 2019 8:57 pm

Hi,Dear Jason:
I did the simulation according to what you said. The peak value is really offset. Thank you. You're the best. I finally know how this place is formed.Thank you very much again!
Attachments
DrTrDOF=False  f=1.1087.png
DrTrDOF=False f=1.1087.png (97.12 KiB) Viewed 2115 times

Ning.Wang
Posts: 17
Joined: Tue Jul 30, 2019 12:52 am
Organization: Ecust
Location: china

Re: load simulation

Postby Ning.Wang » Mon Oct 21, 2019 6:51 am

Hi, Dear Jason:
I just found a strange problem.When the simulation time different,the peak corresponding vaule is different too.The picture in the post is that the simulation time is 100s.The frequency of the peak is about1.1Hz and then look for the Campbell,there is no appropriate frequency vaule,I set simulation time as 600、300s,the frequency vaules of the peak are respectively 1.49、1Hz.The results are as shown in the annex.Also I don't know if it's because of my wind file, Turbsim's wind file is also attached.
I wish you can take a look at it.Thank you very much!
Attachments
TurbSim.inp.png
TurbSim.inp.png (117.93 KiB) Viewed 2103 times
different simulation time DrTrDOF=True&False.png
different simulation time DrTrDOF=True&False.png (149.78 KiB) Viewed 2103 times

Jason.Jonkman
Posts: 4952
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: load simulation

Postby Jason.Jonkman » Mon Oct 21, 2019 10:37 pm

Dear Ning.Wang,

I'm not sure, but I would not normally expect that changing the simulation length would impact the spectra, all else being equal, except for the lowest frequencies, whose resolution would be impacted by the simulation length. But an approx 1.5-Hz mode should be well resolved by a 100- or 600-s simulation. I would guess you've changed other parameters between the simulations

What is the difference between the orange and red curves, and between the blue and green curves? Are these the differences between DrTrDOF = False and True? If so, I don't understand the results because the approx. 1.5-Hz mode is seen for both the orange and red curves. Could the results be labeled incorrectly?

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

Ning.Wang
Posts: 17
Joined: Tue Jul 30, 2019 12:52 am
Organization: Ecust
Location: china

Re: load simulation

Postby Ning.Wang » Tue Oct 22, 2019 1:49 am

Hi,Dear Jason:
I have changed the simulation time as 100s、300s、600s for each simulation,and the same time,3 different simulation times were performed twice under twice under DrTrDOF=true or false respectively,and the picture 1 and 2 can be seen.From thhe results of simulation,we can know that the shift of peak frequency is related to the time of simulation.There is no related to DrTrDOF=true or false.
In addition,the condition of each simulation is changed only one,others are the same.
Attachments
1.png
1.png (158.46 KiB) Viewed 2087 times
2.png
2.png (127.95 KiB) Viewed 2087 times
3.png
3.png (109.46 KiB) Viewed 2087 times

Jason.Jonkman
Posts: 4952
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: load simulation

Postby Jason.Jonkman » Tue Oct 22, 2019 7:11 am

Dear Ning,

I'm sorry, but I don't really believe these results; I would not expect changing the simulation length would change the characteristics of the drivetrain-torsion mode. And I don't see a strong effect from the disabling of the drivetrain-torsion mode. Are you sure there is not an error in the post-processing?

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

Ning.Wang
Posts: 17
Joined: Tue Jul 30, 2019 12:52 am
Organization: Ecust
Location: china

Re: load simulation

Postby Ning.Wang » Tue Oct 22, 2019 8:51 pm

Hi,Dear Jason:
I also can't believe it and I have tried some times but that's the result.I wonder if it has anything to do with the way wind files are computed or run.That's too hard.(cry)
Attachments
turbsim.inp.png
turbsim.inp.png (113.16 KiB) Viewed 2064 times

Ning.Wang
Posts: 17
Joined: Tue Jul 30, 2019 12:52 am
Organization: Ecust
Location: china

Re: load simulation

Postby Ning.Wang » Tue Oct 22, 2019 9:08 pm

Dear Jason:
If there has other possibility that I can try. In addition that the drive chain.Is it possible to have anything to do with 1P? Thank you very much!

Jason.Jonkman
Posts: 4952
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: load simulation

Postby Jason.Jonkman » Wed Oct 23, 2019 1:40 am

Dear Ning,

I do not believe that the results you are showing would be the result of your wind data. I would believe an error in the post-processing.

The 1P frequency is equal to the rotor speed (converted to Hz). I'd also expect to see 2P, 3P, etc. frequencies in the rotating frame and 3P, 6P, etc. frequencies in the fixed frame.

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

Ning.Wang
Posts: 17
Joined: Tue Jul 30, 2019 12:52 am
Organization: Ecust
Location: china

Re: load simulation

Postby Ning.Wang » Wed Oct 23, 2019 7:52 pm

Hi,Dear Jason:
I plan to use other software to da FFT processing in order to verify this possibility.I'll let you know when the results come out.Thank you very much,Jason!


Return to “Structural Analysis”

Who is online

Users browsing this forum: No registered users and 1 guest