Inquiry about RAFT Result Output

[hymdrj]

Hello,

I have a question regarding the output values in RAFT. Could you explain the difference between raft.max_nac_accel and raft.stats_AxRNA? There seems to be a significant discrepancy between the two values. Additionally, is there any documentation available that organizes the various output results that can be extracted from RAFT?

[lucas-carmo]

Hi @hymdrj

I only found max_nac_accel and stats_AxRNA in the code related to OMDAO, which I'm not familiar with. @dzalkind and @mattEhall could you help with that?

Unfortunately, our current documentation doesn't explain the outputs yet. For now, you can get the results directly from a model class object. For instance, considering the example from examples/example_from_yaml.py:

• you can get the mean offsets from model.results['mean_offsets']
• you can get a list of available dynamic responses from model.results['case_metrics'][iCase][iFOWT].keys(), where iCase is the index of the case you want to check (from the list of cases in examples/VolturnUS-S_example.yaml) and iFOWT is the index of the floating wind turbine in the array (this example has only one turbine, so iCase=0). For example, you can get the maximum surge with model.results['case_metrics'][0][0]['surge_max'] (a scalar value) or the power spectral density of surge with model.results['case_metrics'][0][0]['surge_PSD'] (a 1D array with the number of elements equal to the number of frequencies specified by the user)

Hope that helps 😀

[dzalkind]

max_nac_accel should be the maximum over all the cases and stats_AxRNA_max should be the max for each case.

There appears to be a typo, and the standard deviation is used by mistake: https://github.com/WISDEM/RAFT/blob/c9daef3ba24e90372bf481667aee13a7dbed3292/raft/omdao_raft.py#L799

The stats_* outputs should collect all the results that Lucas described.

[hymdrj]

First of all, thank you for your valuable comments. I solved the nac. acc. issue thanks to all.

Also, i have other question about ballast modeling

I’m observing changes by adjusting the ballast grid, but the ballast amount doesn’t change. So, is the ballast amount solely determined by submerged volume and mass? In that case, what is the purpose of defining ballast properties?

Another point I’m curious about is that, based on the code, it seems like the ballast amount is determined based on the input values of the ballast grid.

Lastly, why does the ballast value come out as negative? Could this be due to a mistake in my modeling?

Thank you for reading.

[lucas-carmo]

Ballast properties are computed based on how much of a member is ballasted (specified by l_fill within each member) and on ballast density (rho_fill). I'm not sure what 'ballast grid' is, could you please clarify?

Are you running your case using a .yaml input file? If so, could you please upload your input file here? I can check whether there's something off in your model.