Drivetrain bending in OpenFAST for rotor OMA studies

[Alessandro5I]

Hello,

I am studying Operational Modal Analysis of wind turbines in rotating conditions, with a focus on rotor dynamics.

In OpenFAST, the drivetrain seems to be modelled mainly through rotational DOFs in ElastoDyn. For my work, I would also like to include drivetrain bending dynamics.

I came across the paper “Investigation of main bearing fatigue estimate sensitivity to drivetrain model fidelity in OpenFAST”
https://onlinelibrary.wiley.com/doi/10.1002/we.70005

In that work, drivetrain flexibility is introduced in OpenFAST using SubDyn, while the rotating torsional behaviour is kept in ElastoDyn.

My understanding is that, with this approach, the tower shadow effect is not represented, although it is a very important contribution to the dynamics of the rotating turbine and to rotor-OMA analyses.

So I would like to ask:

What would you recommend to include drivetrain bending in OpenFAST for this kind of analysis?
Is the SubDyn-based approach presented in this paper currently the best option?
Are there any future developments planned in OpenFAST for coupled rotor-drivetrain bending dynamics?

Thank you very much.

[jjonkman]

Dear @Alessandro5I,

The work by Veronica Liverud Krathe (@verlivkra) presents a reasonable workaround in OpenFAST to represent drivetrain degrees of freedom (DOFs) through SubDyn. This approach does currently neglect tower aerodynamic loads and influence (like tower shadow), but we are currently working on improvements to OpenFAST that will enable the aero-elastic coupling between AeroDyn and SubDyn.

We are also working on the coupling between OpenFAST modules and Project Chrono to enable more flexible structural configurations.

We would like to add drivetrain DOFs (like bearing flexibility and shaft bending) directly within ElastoDyn at some point, but we are not currently funded in that area.

Best regards,