Dear Jason,
We’re comparing ElastoDyn and BeamDyn for the one of the standard turbine models (IEA 22MW). For a simulation of duration 1640s, the run times for ElastoDyn and BeamDyn are 3.5 and 400 minutes respectively. This is due to the fact that we are using a very small time step to avoid convergence issues in the BeamDyn solution. ElastoDyn runs fine with a time of 0.05s, but we are forced to use 0.001s with BeamDyn. Do you have any suggestions as to how we might be able to run BeamDyn with a larger timestep? Run times are prohibitively long at the moment.
Appreciate any insights which you may have.
Kind regards,
Aengus.
Dear @Aengus.Connolly,
OpenFAST models with BeamDyn enabled currently run quite a bit slower than desired. We have been addressing this recently through the implementation of a new tight coupling algorithm that should greatly speed up OpenFAST for cases with BeamDyn or SubDyn enabled, by enabling much larger time steps. We expect the release of this updated version in OpenFAST v5.0 later this year. Patience is warranted until then. You can track the progress here: Tight-Coupling Algorithm for OpenFAST 5.0 by deslaughter · Pull Request #2439 · OpenFAST/openfast · GitHub.
Best regards
Dear Jason (@Jason.Jonkman),
Thank you very much for your quick reply (as always). Your insights are much appreciated.
Kind regards,
Aengus.
