Hydrostatic restoring coefficients of Hydrodyn

Hello,

I am currently working on Openfast and more precisely on Hydrodyn and I would like to add hydrostatic restoring coefficients for strip-theory members to AddCLin but I can’t find how to calculate them. The only explanation I found is on the user’s guide document of Hydrodyn but the page’s display is not complete. In the section 6.3.4.8.3. Hydrostatic Restoring for Strip-Theory Members of Floating Systems, I see no equation under the sentence “In its most general form, the 6x6 linear hydrostatic restoring matrix of a floating platform is given by the equation below” :
openfast-wave-stretching.readth … ng-systems
I read some posts on this forum about this matrix, each time the advice given is to refer to the section 6.8.3 of the user’s guide and a link is given but it seems that it’s not available anymore.
If anyone could give me the equation, I would be very grateful.

Best regards,

Dear Julie,

Your link appears to be to a branch of the OpenFAST documentation, which I’m not too familiar with. The complete HydroDyn documentation is available on my Google drive: openfast.readthedocs.io/en/main … index.html).

I’m not sure which version of OpenFAST you are using, but please note that as of OpenFAST v2.6.0 and newer, AddCLin no longer needs to be specified for strip-theory members because the hydrostatic loads on strip-theory members are now recalculated based on substructure displacement.

Best regards,

Hello Sir,

Thank you very much for your quick answer.
I’m currently using the 2.4.0 version so it seems that I need to fill this matrix.
I want to test a potential-flow-only solution and a hydrid solution. The main difference between the both solutions is that hybrid solution takes into account the viscous aspect thanks to the strip theory members, is it right ? It means that in the Hydrodyn input file, I have to put PotMod on TRUE instead of 1, I have to add viscous drag coefficients, as well as strip theory members and joints (that means put some PropPot on FALSE when we want to consider a member as a strip theory member) and I have to add hydrostatic restoring coefficients for strip theory members. Am I forgetting something ?
I have a last question concerning the hydrostatic restoring matrix. For the potential flow theory, I need file .1, .3, and .hst of WAMIT. I don’t have access to WAMIT, so I wrote file .1 and .3 thanks to another Hydrodynamic software. Concerning file .hst, that contains the hydrostatic and gravitational restoring coefficients given here in section 4.1: https://www.wamit.com/manual6.4/Chap4.pdf, I read that the mass restoring shouldn’t be taken into account therfore the three coordinates of the center of gravity should be put equal to 0. However to calculate the integrals, I need to know how they define what they call the « mean body wetted surface, Sb ». At first sight I thought that the surface considered was the waterplane area, like in a classic definition of hydrodynamic restoring coefficients. But if I take the waterplane area for Sb, it seems that some terms are going to be double couting in hybrid solution, according to Section 6.8.3.

Best regards,

Dear Julie,

Yes that sounds correct, except PotMod is a switch, not a flag (valid values of PotMod are 0 = no potential flow, 1 = based on WAMIT out, and 2 = based on fluid-impulse theory). To clarify:

  • For a potential-flow only solution, set PotMod = 1 and you don’t need to specify any strip-theory joints, members, or any coefficients.
  • For strip-theory only solution, Set PotMod = 0 and specify the strip-theory joints, members and all coefficients.
  • For a hybrid potential-flow plus strip-theory solution, set PotMod = 1 and specify the strip-theory joints, members, and drag coefficients (no need to specify added mass or pressure coefficients) and set PropPot = True for the members that you want to model with the hybrid formulation; this will add viscous drag to the potential-flow solution for those members.

Regarding the hydrostastic restoring matrix, I agree that the body weight term should not be included because this is already captured in the ElastoDyn and SubDyn structural modules of OpenFAST. The integrals over S_b in the WAMIT manual are the integrals over the full wetted surface of the body in the undisplaced condition (not just the waterplane area). In a hybrid formulation, all terms except viscous drag are accounted for in the potential-flow solution, so, there is no need to specify AddCLin in that case (the values from the *.hst file are used instead). If some strip-theory members are not accounted for in the potential-flow solution, then you’d want to specify the AddCLin matrix for those members in OpenFAST v2.4, but this is not needed in OpenFAST v2.6 and newer.

I hope that helps.

Best regards,

Hello,

After I read this thread, I got a question regarding the AddClin. I am using FASTv.8.16. I checked the additional hydrostatic stiffness in the Hydrodyn manual and also check the test file from NREL. I check the OC4DeepCWind Semi model. In the hydrodyn of this model, I saw that you input only element 44 and 55. For the other elements such as the off-diagonal elements, I understood why they are zero. But for element 33 (RhogA0), I don’t understand why you did not input it? Can you please explain this??

Dear Danupon,

The nonzero AddCLin matrix for the OC4-DeepCwind semisumersible model reflects only the restoring associated with total mass of ballasting/flooding (m_f in the HydroDyn manual); only the (4,4) and (5,5) elements are nonzero in the model pre OpenFAST v2.6.0. The remainder of the hydrostatic restoring is captured in the potential-flow solution (WAMIT *.hst file).

Best regards,