Dear @Peng.Guo,
Good to hear that worked for you! Those changes were added to the SoilDyn branch about 10 months ago – we made some architectural changes to how SubDyn meshes worked with flexible floating platforms then. Not having the y3Mesh change would definitely cause issues with the model stability.
Did you run into any conflicts when merging the latest main into the SoilDyn branch?
Regards,
Andy
Dear @Andy.Platt
I could run the exe file smoothly.
However, I still have some questions to ask you.
I have added codes for subroutine SlD_CalcOutput to calculate damping force in SubDyn.f90. and I also improved SoilDyn_Registry.txt and SoilDyn_IO.f90.
I’m not sure if those codes are problematic. I’d appreciate it if you could check it out for me.
Dear @Jason.Jonkman and @Andy.Platt ,
I am using the SoilDyn source code to model the pile-soil interaction. What I want to know is whether to treat the extension section of the tower towards the mudline as a single pile when using distributed spring model (DS model)? Thanks in advance.
Best regards,
Kevin
Dear @Kevin.Chang,
I’m not sure I understand your question. A monopile would have a single pile whereas a jacket foundation would have multiple piles.
Best regards,
Dear @Jason.Jonkman ,
I’m sorry for not stating it clearly. For a monopile-supported offshore wind turbine, how to model a monopile in OpenFAST (including the SoilDyn source code)? Whether the monpile is modelled by using the method of extending the tower downwards from the mudline?
Best regards,
Kevin
Dear @Kevin.Chang.
My understanding is that the SoilDyn module has a placeholder for a distributed springs (e.g., P-Y curve) model, but that this feature has not been fully developed. If you were to use a distributed springs model (not available without a source code change), you would have to model the pile in SubDyn as a structural member to its free end below the seabed. However, SoilDyn only currently supports a coupled springs representation of the foundation, including through simple linear matrices or through coupling to the REDWIN superelement. To model a coupled springs foundation, you should only model hte pile in SubDyn as a structural member down to the seabed.
Best regards,
Although not ideal, I understand that one option to include linear distributed springs (i.e., different linear p-y relationships along the pile) would be working with SubDyn. In this case, you should include in SubDyn the monopile (including the part below the seabed) and reaction joints with associated stiffness matrices.
Kind regards.
Hello everyone,
I know that this question has been asked a lot. But I want to make sure I understand correct.
I want to simulate the soil structure interaction of monopile offshore wind turbine, I want to include NONLINEAR stiffness matrix at the seabed
I know that OpenFAST can include LINEAR stiffness matrix in SubDyn through SSI file.
I red that SoilDyn module can include the nonlinear soil structure interaction, but is not ready yet.
Can I use FAST v7 to include NONLINEAR stiffness matrix through UserPtfmLd_CS.f90.txt? or this method is limited to LINEAR stiffness matrix only? If it can include NONLINEAR stiffness matrix at the seabed, where can I find the FAST v7 files for NREL 5 MW monopile offshore wind turbine?
Regards,
Marwa
Dear @Marwa.Mohamed,
Your understanding is correct. In OpenFAST, you could implement the nonlinear stiffness matrix for a monopile offshore wind turbine through the SoilDyn module. In FAST v7, you could implement the nonlinear stiffness matrix for a monopile offshore wind turbine through the UserPtfmLd_CS.f90 routine. Given that you’ll have to change the source code and recompile either way, I would suggest choosing OpenFAST rather than downgrading to the much older FAST v7. You can find the current branch of SoilDyn in the following PR: SoilDyn -- soil dynamics module by andrew-platt · Pull Request #986 · OpenFAST/openfast · GitHub. If your nonlinear stiffness includes both linear and nonlinear terms, I would recommend including the linear contribution within the SSI capability of SubDyn, and only add the nonlinear residual through SoilDyn.
Best regards,
Thanks a lot for your help.
Dear Dr Jason,
1-Does UserPtfmLd_CS.f90 routine can include NONLINEAR stiffness matrix and do iterations internally to consider stiffness matrix and update the loads iteratively inside the FAST 7 or not?
2- if Yes, is there any solved example for these files please?
3- How FAST 7 decide that the inputted stiffness matrix through UserPtfmLd_CS.f90 routine linear or nonlinear?
Regards,
Marwa
Dear @Marwa.Mohamed,
Here are my responses:
UserPtfmLd() routine is a placeholder for user-defined platform reaction loads in six DOF. The UserPtfmLd_CS.f90 example I’ve shared on this forum in the past implements a linear coupled springs matrix, but you could change this to be nonlinear if you want. FAST v7 does not have anything comparable to OpenFAST in terms of over-loop correction steps.UserPtfmLd() routine with nonlinearity to share as an example.UserPtfmLd() routine includes nonlinearities or not. It is up to you to implement the UserPtfmLd() routine as you see fit.All that said, again I’d recommend using OpenFAST over FAST v7 given that OpenFAST is superior to FAST v7 in most ways.
Best regards,