Clarification on Coordinate Systems: Relationship between PtfmRef (HydroDyn) and PtfmCM (ElastoDyn) for Asymmetric Platforms

[jmmungod6]

Hi,

I am currently working on a simulation for an asymmetric semi-submersible floating platform (similar to a WindFloat concept), where the wind turbine is located on one of the outer columns rather than the geometric center.

After reviewing the documentation and similar issues, I am still encountering some uncertainties regarding the correct definition of the coordinate systems, specifically the interaction between PtfmRefxt/yt/zt in HydroDyn and PtfmCMxt/yt/zt in ElastoDyn.

I am observing large, unrealistic oscillations (instability) in the platform's motion, which leads me to believe I may have a mismatch in how I am mapping the WAMIT solution to the OpenFAST structural model. I would appreciate it if you could confirm if my understanding of the following two points is correct:

1. The definition of PtfmRef in HydroDynMy
understanding is that PtfmRef acts as a translation vector that locates the origin of the WAMIT solution (Potential Flow body reference) with respect to the OpenFAST Global Inertial Frame (0,0,0 at MSL/Tower Center).Does this imply that the following two workflows are mathematically equivalent?

• Workflow A: I mesh the platform in WAMIT such that the tower centerline is at (0,0,0). In this case, since the WAMIT origin coincides with the OpenFAST global origin, I should set:PtfmRefxt, PtfmRefyt, PtfmRefzt = 0, 0, 0.

• Workflow B: I mesh the platform in WAMIT using its geometric center (e.g., $X_{geo} = 40$m from the tower). In OpenFAST, to align the physics correctly, I should set:PtfmRefxt = 40.0 (The distance from OpenFAST (0,0,0) to WAMIT (0,0,0)).

Is PtfmRef solely responsible for shifting the application point of the diffraction/radiation forces and the added mass matrices calculated by WAMIT?

2. The reference frame for PtfmCM in ElastoDyn

This is where I have the most doubt. How is PtfmCM (Platform Center of Mass) defined in relation to PtfmRef? Is PtfmCMxt defined relative to the OpenFAST Global Origin (0,0,0)? Or, is PtfmCMxt defined relative to the HydroDyn PtfmRef point?

For an asymmetric platform where the mass is centered at $X=45$m, but the WAMIT diffraction analysis was performed at $X=0$ (Tower), should PtfmCMxt be entered as 45.0, regardless of what PtfmRefxt is set to?

Any clarification on whether these two parameters are independent vectors from the global origin or if they are chained would be incredibly helpful to solve the instability issues I am facing.

Thank you very much for your time and support.
Best regards,

[jjonkman]

Dear @jmmungod6,

Regarding (1), I agree with your interpretation of PtfmRefxt, PtfmRefyt, and PtfmRefzt in HydroDyn.

Regarding (2), I agree with your understand of PtfmCMxt and PtfmRefxt. Both of these are defined with respect to the same origin, not relative to each other. They can also be independent from the platform reference point(s) in HydroDyn.

FYI: I would say the the origin in HydroDyn and ElastoDyn are not "global", but rather body-fixed within the floater and only align with the global inertial frame origin if the platform displacements are zero.

Best regards,

[jmmungod6]

Dear @jjonkman

Thank you very much for your quick and clear response.

It was extremely helpful, especially the clarification that the ElastoDyn and HydroDyn inputs are defined with respect to the same origin (body-fixed) and are not relative to each other. I have now tested both workflows mentioned in my previous question (Method A and B) with the correct coordinate definitions, and both are working perfectly without the instability I was seeing before.

Best regards,