search

OpenFAST / openfast

Main repository for the NREL-supported OpenFAST whole-turbine and FAST.Farm wind farm simulation codes.
http://openfast.readthedocs.io
Apache License 2.0
671 stars 452 forks source link

about Moordyn #2326

Closed HANloading closed 1 month ago

HANloading commented 1 month ago

Dear All, I have some questions about the operation of Moordyn, I am considering a semi-submersible platform with three columns, each column is connected to 3 anchor chain lines, so there are 9 anchor chain lines in total, and I would like to ask if moordyn can handle 9 lines? If not, what is the maximum number of lines that can be handled? If it can, does the value of BA/-zeta affect the accuracy of the anchor chain load? Or are there other factors that can affect the accuracy of the anchor chain load? Best regards! image

RyanDavies19 commented 1 month ago

Hi @HANloading,

There is no limit to the number of lines that MoorDyn can handle, but with more lines comes increased computation times. BA/-zeta is the value for the internal damping of the line which helps with numerical stability by reducing high frequency vibrations. Typically it is set to -1. For the anchor chain loads, the values you get will be depend on the chain properties you provide in the MoorDyn input file (see input files in the docs). MoorDyn's modeling of lines has been well validated, provided you give the correct input parameters for the lines you are trying to model.

HANloading commented 1 month ago

@RyanDavies19 Thank you for your answer. The problem I'm having now is that when I run moordyn, it prompts some warning messages, and I'm not really sure what's going on, and when this load data is applied to my model, the model's platform produces unpredictable movements. If I know the pre-tension of each anchor chain line, in addition to the known equivalent mass, the vertical combined force. Can I get the anchor chain static matrix? image image Lines.txt

RyanDavies19 commented 1 month ago

@HANloading

It's hard to say what's happening without knowing the MoorDyn errors or the bigger simulation you are running. However based on your input file I cold see some potential issues. First your mooring lines are 750m long, however your fairlead and anchor points are further apart than that (example point 1 and 2 have horizontal difference of 750m and vertical difference of 70m). If that doesn't crash MoorDyn it will cause a very taut mooring system. Additionally, your water depth is 200m, while the deepest point you have is only at -55m. Lastly it seems you switched fairlead and anchor points after point 10, because your coupled points are now at -55m and your fixed points are at +16m.

With regards to the quasi-static mooring stiffness matrix, that is not a feature of MoorDyn because it is a dynamic model. If you need the stiffness matrix refer to the OpenFAST forum for information: https://forums.nrel.gov/search?q=stiffness%20matrix%20mooring

HANloading commented 1 month ago

Dear @RyanDavies19 Now that the water depth and the unstretched length of the anchor chain line have been modified according to tender's suggestion, although it can run successfully and express the anchor chain load more correctly, running the device prompts some warnings, may I ask the reason for this warning why exactly? image

RyanDavies19 commented 1 month ago

@HANloading

I believe that warning just means the catenary solver failed and so MoorDyn will solve the line profile using dynamic relaxation. If it converges during IC generation to a solution then you are fine. MoorDyn process for finding the steady state initial conditions of the system is to first attempt to solve a quasi-static approach to the catenary shape. If that fails it gives the line a linear profile. From there it allows the line to 'fall' into place with scaled drag coefficients until the fairlead tension is steady. Once that occurs, or the IC solver reaches the TmaxIC threshold, the simulation begins.

However, it looks like you are using a very old version of MoorDyn/OpenFAST that has not been updated in 7 years. If possible please try to use the most recent release of the code: https://github.com/OpenFAST/openfast/releases/tag/v3.5.3

For transferring your MoorDyn v1 input file to MoorDyn v2 format, the documentation has descriptions of both file types that should make it possible to convert things over: https://moordyn.readthedocs.io/en/latest/inputs.html

HANloading commented 1 month ago

Dear @RyanDavies19 Thank you very much for your reply, it has successfully solved my problem!