Dear Jason
I am a newcomer to FAST. And I wonder since there already is seismic in FAST.v7, why OpenFAST don’t have the ability to calculate earthquake. Or , I mean, I wonder what makes it not easy to implement seismic from v7 to OpenFAST, with seismic related code already been coded, and OpenFAST actually developed from v7?
Thank you! Best wishes.
Dear @KeHao.Qian,
The short answer is that NREL has not yet been funded to port the FAST v7 functionality into OpenFAST. There are also a few changes we’d like to see made the Seismic functionality of FAST v7 before porting to OpenFAST that prevent this from being a trivial task.
That said, you can mimic what the Seismic functionality of FAST v7 does via the external platform MCKF (CompSub = 2) option of OpenFAST, as has been discussed in several forum topics.
Best regards,
1 Like
Dear @Jason.Jonkman ,
I’m a student learning OpenFAST for combining the seismic, ice and wind load and applying in wind turbine. So i want to ask you if there has a method to achieve seismic load in OpenFAST?
Thank you! Best wishes.
Dear @Jack.Hardy,
My answer is the same as I gave KeHao.Qian. The external platform MCKF (CompSub = 2) option of OpenFAST enables you to prescribe your own platform motion (as has been discussed in other forum topics), but the full Seismic functionality of FAST v7 has not yet been ported into OpenFAST.
Best regards,
Dear @Jason.Jonkman ,
Thank you for your answer! But i can’t find where is MCKF(CompSub=2).Can you tell me its location?
Best wishes!
Dear @Jason.Jonkman ,
I already find the example about ExtPltm.But the details about input files can’t be find in OpenFAST documention?
Dear @Jack.Hardy,
Documentation on the external platform MCKF (CompSub = 2) option of OpenFAST is available in OpenFAST readthedocs documentation here: 4.2.6. ExtPtfm — OpenFAST v3.5.3 documentation.
Best regards,
Dear @Jason.Jonkman,
Thank you for your answer! I already read the documention and find the files about ExtPtfm.But I find there only ‘compElast’ and ‘compSub’ keep activative in the input file ‘5MW_OC4Jckt_ExtPtfm.fst’.So the first i want to ask you if i can activate ‘compInflow’ , ‘compAero’ and so on?
Then i find the file ‘ExtPtfm_SE.dat’ has 31 dimensions, i don’t know how to define the dimensions and input Seismic acceleration time history . I think the seisimic load achieveing by the ‘loading and wave elevation’,is right?
i hope you can guide me about the above questions.
Best wishes!
Dear @Jack.Hardy,
Yes, you can enable aerodynamics, etc. in an OpenFAST model with CompSub = 2.
The r-test example with CompSub = 2 that you are referencing has 6 Guyan modes + 25 Craig-Bampton modes, for a total of 31 DOFs, which is why the matrices are size 31. To model seismic loading following the approach in the old Seismic functionality of OpenFAST, you can disable all Craig-Bampton modes. In this case, CombSub = 2 should provide the 6x6 mass, damping, and stiffness matrices and 6x1 load vector for the Guyan modes. I would following the documentation from the Seismic functionality (https://www.nrel.gov/wind/nwtc/assets/pdfs/sr-5000-53872.pdf) to derive these matrices and load vector.
Best regards,
thanks for your guide.
Dear @Jason.Jonkman ，
I read many archives about ExtPtfm. I have some thinks need you to help me modifying. In fact,these are my reasoning.
This is how i understand the matrix that the numerical values of mass matrix are much larger than the real mass of the turbine, the numerical of stiffness matrix are calculated by this equation,
and damping matrix is that
,the force calculation equation is
but i really can’t understand the ‘PtfmDisp is the desired displacement at the base of the tower; X is the realized displacement’,i know the desired displacement is the seismic acting,but the realized displacement is what? I don’t konw if my thinks about mass,damping,stiffness matix are right?
Last,I want to if ExtPtfm can simulate the seismic action of land turbine? Because i’m trying to reaserch the dynamic response of land turbine under multi-load.
I’m sorry to bother you.
Best regards.
Dear @Jack.Hardy,
I agree that the mass, damping, and stiffness of ExtPtfm should be defined by these equations from the Seismic functionality documentation (at least for the transnational DOFs). The X and XD terms will be calculated by OpenFAST as part of the solve when you specify k and c as the stiffness and damping terms in ExtPtfm. The force specified in ExtPtfm should be equal to k * PtfmDisp + c * PtfmVel.
Best regards,
thank you for your answer！
Dear @Jason.Jonkman ,
I’m trying to get the biggest frequency of the wind turbine.I use Bmodes to analyse ‘5MW_Land_DLL_WTurb’ files. But the frequency from Bmodes is very different from the NERL file.
And the ‘.bmi’ and “tower_secs.dat” files are following:
Dear @Jack.Hardy,
Can you clarify what you mean by ‘5MW_Land_DLL_WTurb’ files in regards to BModes?
The Table 9-1 you are referring to is for the land-based version of the NREL 5-MW baseline wind turbine, but I see that you’ve defined a free-free boundary condition in BModes (hub_conn = 2), which is why the first six frequencies are very low (rigid-body modes). The tower-base should be clamped for a land-based wind turbine, so, I would suggest using hub_conn = 0 in BModes.
You mention your interest in getting the “biggest frequency”; I’m not sure why you mean by that.
Best regards,
Dear @Jason.Jonkman ,
Thank you for your reply.The biggest frequency is mean the seismic parameter ‘Actfreq’ . The document suggest it’s value should be at least 10 multiply the biggest frequency of strcuture ?
Best regards.
Dear @Jason.Jonkman ,
The ‘hub_conn’ could be 1,2,or3,can’t be 0. So i define the value be 1.But the result still not expected.
Best regards.
Dear @Jack.Hardy,
Oops, I meant hub_conn = 1. With hub_conn = 1, your tower frequencies calculated by BModes are similar to the published report – that is 0.34 Hz from BModes instead of 0.32 Hz from the publication. It looks like you are now running BModes correctly.
Regardless, what you want to calculate ActFreq from is the highest frequency of the OpenFAST model, not of a BModes model. You can obtain the highest frequency of an OpenFAST model via linearization analysis, followed by eigenanalysis, which has been discussed in other forum topics.
That said, I know that the highest frequency of the OpenFAST model of the land-based NREL 5-MW baseline wind turbine is around 6 Hz, so, you could use ActFreq = 60 Hz per the Seismic rule of thumb you are referring to.
Best regards,
Thanks for your reply.
Dear @Jason.Jonkman ，
I’m so sorry to bother you so many times.But i really have many questions about OpenFAST.Now I’m trying to establish a tower model by myself,but the tower properties i don’t know the calculate method.
I don’t know the stiffness about every paragragh how to calculate. For example,the fisrt paragragh if it means from 0.0m to 8.76m ? i think it’s wrong,it maybe from 0.0m to a height , so the hight is how to define? and the stiffness is how to calculate?
Best regards.