Thank you @Jason.Jonkman. Regarding servodyn, can I just simply disable servodyn or change some settings in the servodyn input file for the parked-feathered condition?
Dear @Rahul.Ramachandran,
You can disable ServoDyn when simulating the parked/idling condition if you want to consider fixed pitch and fixed yaw with YawDOF = FALSE.
Best regards,
Hello Jason,
Is it possible to simulate the motion of a floating platform (without the wind turbine) using openfast ? I tried changing all the masses and inertias of the tower and RNA, mass/unit length of tower and blade segments to zero and when I run openfast, I get an error. I tried using the hydrodyn standalone version also but the platforms motions are not calculated. Where do I input the inertia and mass details of the platform, as an elastodyn file is not an input for the standalone hydrodyn module ? Please tell me if this is the right approach.
Regards,
Rahul.
Dear @Rahul.Ramachandran,
What error did you get?
You can’t eliminate the blades, drivetrain, nacelle, and tower from an OpenFAST model, but you can set the inputs so that terms associated with the blades, drivetrain, naccelle, and tower are zero or negligible. For example, you should disable wind inflow, aerodynamics, and control; disable the blade, drivetrain, and tower degrees of freedom in ElastoDyn; set the geometry so that the blade and tower lengths are zero; and set the masses and inertias to zero or negligible values, etc.
As you said, the standalone HydroDyn driver does not calculate platform motions; rather motions are prescribed when using the driver. As such, the platform mass and inertia details are not specified in this driver because they are not needed for the calculation.
Best regards,
Hi Jason,
Thank you so much.
I changed the BMassDen in the Blade Elastodyn file to zero and I got this error:
Dear @Rahul.Ramachandran,
There are some input checks in ElastoDyn that prevent unphysical inputs and/or trap errors before a simulation crash. ElastoDyn will not allow you to set BMassDen exactly zero, but you can set BMassDen very close to zero (e.g., 0.000001) to avoid the error. This is what I was referring to when I mentioned “negligible” values.
Best regards,
Hi all,
I’m writing because I’m using NREL’s SOWFA coupled with OpenFAST to simulate via LES a FOWT subject to prescribed sinusoidal surge and sway motions. At this stage I’m more interested in seeing the effect these motions have on the wake dynamics and power performance rather than on the wind turbine structure and aerodynamic loads, although I don’t rule out analyzing them in the future. Furthermore, the characteristics of the substructure or even its mere presence are not of my interest. I’m only interested in having a translation of the whole machine (rotor, nacelle and tower).
Based on the following threads: Regarding Seismic Load by FAST code and https://forums.nrel.gov/t/how-to-input-platform-motions-time-series-and-observe-the-responses/2026, it seems that the Seismic module may be used to specify a platform motion time series independently in all three axes. However, it’s mentioned by @Danupon.Subanapong in the latter that this module cannot be coupled with OpenFAST. Is this true? I’ve read in this thread that @Ricardo.Amaral has used the ExtPtfm module to prescribe such motions (see https://forums.nrel.gov/t/rigid-wind-turbine-with-harmonic-motions/685/13?u=juan.monmeneu). However, it seems that the operation of this module is different.
If I’m not mistaken, one has to perform a time-domain simulation of the isolated substructure using an external program to produce a reduced model. The results of this simulation are written to a file containing the system matrices and the time series of the interface reduced input loads, which are then used by ExtPtfm to perform a simulation in time using this reduced model. Finally, OpenFAST outputs times series of modal loads and displacements at the interface point (correct me if I’m wrong).
What would you recommend I do to prescribe surge and sway motions to the whole structure in OpenFAST without having to perform additional simulations for the substructure?
Thanks in advance
Kind regards,
Juan
Dear @Juan.Monmeneu,
I agree that NREL has never coupled the Seismic functionality of FAST v7 to OpenFAST.
The ExtPtfm module of OpenFAST was developed to support a superelement representation of the support structure, as you summarize in your paragraph:
If I’m not mistaken, one has to perform a time-domain simulation of the isolated substructure using an external program to produce a reduced model. The results of this simulation are written to a file containing the system matrices and the time series of the interface reduced input loads, which are then used by ExtPtfm to perform a simulation in time using this reduced model. Finally, OpenFAST outputs times series of modal loads and displacements at the interface point (correct me if I’m wrong).
However, you can use ExtPtfm in a simplified way to prescribe the 6-DOF motion of the platform supporting the wind turbine, similar to how the Seismic module of FAST v7 functioned. See the following topic on OpenFAST GitHub Discussions for more information: Develop Prescribed Platform Motion capability with aero-elasticity · OpenFAST/openfast · Discussion #872 · GitHub (particularly my post dated Dec 23, 2020).
Best regards,