Simulating the parked turbine

Provide feedback, request enhancements, and get help with wind-turbine computer-aided engineering tools.

Moderators: Bonnie.Jonkman, Jason.Jonkman

Jason.Jonkman
Posts: 5069
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: Simulating the parked turbine

Postby Jason.Jonkman » Tue Nov 26, 2019 10:24 am

Dear Salem,

Sorry, but I'm still not sure I know what the cause of this numerical instability is. I also see a spike in the rotor speed. What happens if you disable the controller (PCMode = 0) and fix the rotor speed at rated (GenDOF = False, RotSpeed = 12.1 rpm)?

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

Lixian.Zhang
Posts: 18
Joined: Wed Dec 05, 2018 11:04 pm
Organization: Dalian University of technology
Location: China

Re: Simulating the parked turbine

Postby Lixian.Zhang » Fri Dec 27, 2019 4:20 am

Dear Dr. Jason
Thanks for you reading my message.
I'm using FAST V8 to simulate the parked condition of floating wind turbine in shallow water depth (50 m). And FAST stop running because of large platform motion. I have searched another people question here which is similar to my problem. In the load case defination, the extreme sea condition is Hs=9.5 m and Tp is 17.28 s. I think maybe the reason is that the natural frequency of the platform pitch motion (19 s) is close to the peak period. Therefore, I added additional damping in pitch motion, and used AeroDyn v14 to simulate the aerodynamic loads. It worked eventually.
1'. My frirst question is that why AeroDyn v15 can not work even if I set the wakemode=0 and AFAeroMod = 1. Note: both for the use of AeroDyn v14 and v 15, The RotSpeed, GenDOF, PCMod were all changed to 0, and BlPitch are changed to 90 degree.

2. Even if AreoDyn v14 can worked, I find the simulation results is weird. As we can see in the following pictures. Is the thrust force normal in parked condition? Why the negative value appeared? And I think pitch motion is too large in parked condition.
推力.png
推力.png (71.87 KiB) Viewed 1966 times
Pitch.png
Pitch.png (65.13 KiB) Viewed 1966 times


I would be appericate if you can give me some advice!
Thank you very much!

Best regards
Lixian Zhang

Jason.Jonkman
Posts: 5069
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: Simulating the parked turbine

Postby Jason.Jonkman » Sat Dec 28, 2019 9:03 am

Dear Lixian,

Regarding (1), I'm not sure what you mean when you say "why AeroDyn v15 can not work even if I set the wakemode=0 and AFAeroMod = 1". I would expect that AeroDyn v15 would work without problem with these settings and give the same results as AeroDyn v14 under equivalent settings. What problem are you running into?

Regarding (2), you haven't stated which FAST output you are plotting, but I would assume you are using ElastoDyn output RotThrust. This output is the reaction force transmitted from the rotor to the shaft, so, not only includes the applied aerodynamic forces, but also the forces resulting from rotor weight and inertia (see related forum posts on this topic), which is likely why the thrust swaps sign. You haven't stated what floating system you are modeling, so, I can't comment on if the platform-pitch motion is appropriate for the conditions simulated.

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

Lixian.Zhang
Posts: 18
Joined: Wed Dec 05, 2018 11:04 pm
Organization: Dalian University of technology
Location: China

Re: Simulating the parked turbine

Postby Lixian.Zhang » Mon Dec 30, 2019 4:40 am

Dear Dr.Jason

Thanks very much for your quick response.
Regarding (1) : thank you, I have solved the problem, AeroDyn v15 can work now.
Regarding (2): Yes, I am using elastoDyn output Thrust force. After I checked the related forum post, I understood that the RotThrust is not only including the aerodynamic loads. I also found RtAeroFxh can output pure aerodynamic loads according to the forum post and the aerodynamic loads under extreme sea condition is small. The Floating system used in the simulation is V-shaped semisubemersible FOWT as shown below. The pitch natural frequency is around 19 s and peak frequency of the wave spectrum is 17.02 s. Will it be the similar period between pitch mode and wave frequency that causes the excessive structural response in pitch motion? And is the platform-pitch motion appropriate for the conditions simulated?
V-shapedsmei.png
V-shapedsmei.png (40.77 KiB) Viewed 1950 times


Thanks again!


Best regards
Lixian Zhang

Jason.Jonkman
Posts: 5069
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: Simulating the parked turbine

Postby Jason.Jonkman » Mon Dec 30, 2019 7:15 am

Dear Lixian,

Yes, I would guess the closeness between the wave-excitation frequencies and the platform-pitch natural frequency is the cause for the large pitch motion in your case. If you haven't already, I would suggest calculating the Response Amplitude Operators (RAOs) of the system to better understand its behavior. You can calculate RAOs using FAST, as discussed in our ISOPE 2013 paper: https://www.nrel.gov/docs/fy13osti/58098.pdf.

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

Lixian.Zhang
Posts: 18
Joined: Wed Dec 05, 2018 11:04 pm
Organization: Dalian University of technology
Location: China

Re: Simulating the parked turbine

Postby Lixian.Zhang » Tue Dec 31, 2019 4:03 am

Dear Dr.Jason

Thanks for you suggestions! Yes, maybe I should calculate the RAO first.


Best regards
Lixian Zhang

Salem.Okpokparoro
Posts: 22
Joined: Fri Feb 16, 2018 8:15 am
Organization: University of Aberdeen
Location: United Kingdom

Re: Simulating the parked turbine

Postby Salem.Okpokparoro » Fri Jan 31, 2020 7:22 am

Dear Jason
I am still unable to simulate 50 years extreme conditions of Uref 51m/s, Hs=13m and Tp=14s for the OC3-Hywind. The simulation keeps getting numerically unstable with the Small angle assumption violated in SUBROUTINE SmllRotTrans() due to a large platform displacement. Setting PtfmPDOF= False makes the simulation run without errors. I have only changed the environmental conditions in the Test24 of FAST repository.
I look forward to getting your invaluable suggestions on how to address this issue.
Regards
Salem

Jason.Jonkman
Posts: 5069
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: Simulating the parked turbine

Postby Jason.Jonkman » Fri Jan 31, 2020 9:27 am

Dear Salem,

Earlier in this forum topic you were asking questions about DLC 1.6a, where the turbine is operating under extreme waves. Did you finally get that simulation to work as expected?

Are you now asking about load case 6.X, where the turbine is parked/idling under extreme waves? Have you pitched the blades to feather, disabled the torque, and disabled the aerodynamic induction and unsteady airfoil aerodynamics?

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

Salem.Okpokparoro
Posts: 22
Joined: Fri Feb 16, 2018 8:15 am
Organization: University of Aberdeen
Location: United Kingdom

Re: Simulating the parked turbine

Postby Salem.Okpokparoro » Fri Jan 31, 2020 9:40 am

Dear Jason
I am still unable to run simulations with extreme sea state either Dlc1. 6 or 6. x. For the latter the idling condition has been set up : blades feathered at 90deg, aerodyn setting set as you've indicated on several posts, pcmode disabled. The simulations go unstable with the sea state defined by the 50yr Hs and Tp. I have tried both FASTv8 and OpenFAST. When i reduce HS to around 10m the simulations run well or if i disable Platform pitch dof.
Regards

Jason.Jonkman
Posts: 5069
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: Simulating the parked turbine

Postby Jason.Jonkman » Fri Jan 31, 2020 9:53 am

Dear Salem,

For DLC 1.6a, did you try disabling the controller (PCMode = 0) and fixing the rotor speed at rated (GenDOF = False, RotSpeed = 12.1 rpm) as I asked on Nov 26?

For DLC 6.X, what does the response look like when you say it is unstable? Does increasing the yaw spring stiffness have any effect on DLC 6.X?

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

Salem.Okpokparoro
Posts: 22
Joined: Fri Feb 16, 2018 8:15 am
Organization: University of Aberdeen
Location: United Kingdom

Re: Simulating the parked turbine

Postby Salem.Okpokparoro » Sat Feb 01, 2020 1:38 pm

Dear Jason
The responses for DLC6.x are attached. The simulation still goes unstable even when the yaw stiffness is increased by x1000.
Regards
Attachments
responses.jpg
responses.jpg (87.32 KiB) Viewed 1653 times

Jason.Jonkman
Posts: 5069
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: Simulating the parked turbine

Postby Jason.Jonkman » Sat Feb 01, 2020 7:24 pm

Dear Salem,

Is the large surge motion caused from tower aerodynamic drag? (I would expect the rotor thrust to be small when the blades are feathered to 90 deg.)

My guess is the system was not designed for the severity of these conditions and the applied forces exceed those able to restore the spar back vertically.

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

Salem.Okpokparoro
Posts: 22
Joined: Fri Feb 16, 2018 8:15 am
Organization: University of Aberdeen
Location: United Kingdom

Re: Simulating the parked turbine

Postby Salem.Okpokparoro » Sat Feb 29, 2020 2:08 am

Dear Jason
Same errors occur when simulating dlc1. 6a with lesser tower drag. How can I simulate conditions with Hs>=10m for the spar? When i reduce the Hs to <9m the simulations run smoothly. But these are less than the 50yrs Hs of the reference site. Although these large waves induce large platform motions, they are still physically realistic. Any advice would be greatly appreciated.
Regards
Salem

Jason.Jonkman
Posts: 5069
Joined: Thu Nov 03, 2005 4:38 pm
Location: Boulder, CO
Contact:

Re: Simulating the parked turbine

Postby Jason.Jonkman » Sat Feb 29, 2020 8:45 am

Dear Salem,

I'm still guessing that the spar system you are analyzing was not designed for the severity of the conditions you are applying, but I have not ran into this problem myself, and so, have not studied it in detail. It may be interesting to assess the stiffness and damping characteristics of the spar at the large floater surge/pitch (etc.) offset just before simulation crash. Could the stiffness be dropping considerably or does the damping go negative? You could test this with the new floating offshore linearization capability we've recently added to OpenFAST--see: https://www.nrel.gov/docs/fy19osti/71865.pdf and https://iopscience.iop.org/article/10.1 ... 6/1/012022. While not yet included in the master or dev branches of OpenFAST, it is available in a pull request (just awaiting final review and approval before merged into dev): https://github.com/OpenFAST/openfast/pull/350.

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

Chenxu.Zhao
Posts: 11
Joined: Mon May 20, 2019 4:36 am
Organization: Chongqing University
Location: China

Re: Simulating the parked turbine

Postby Chenxu.Zhao » Sun May 03, 2020 2:05 am

Dear Jason:
I’m running a test of parked 5MW onshore wind turbine. I set the tower’s DOF and blade’s DOF to be “Ture”.It can be seen from the calculation results that the bending moment at the bottom of the tower in the side-side direction is 4200 kN·m and the mean concentrated bending moment at the hub is about 670 kN·m. Eccentric bending moment due to gravity is about 1388kN·m. so the mean bending moment caused by mean lift force should be 4200-1388-670=2142 kN·m(Since the lift force on the tower is almost 0, I ignored it).
But I find the mean lift force of rotor is about 18kN. The hub height is about 90m.so the mean bending moment caused by mean lift force should be 18×90=1620 kN·m.
Why the error reaches about 20%?is there any other bending moment I missed?


Return to “Computer-Aided Engineering Software Tools”

Who is online

Users browsing this forum: No registered users and 1 guest