Compute surface elevation using IFFT #250
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Previously providing a zero frequency when creating a spectrum (JONSWAP or Pierson Moskowitz) produced a NaN value. This is due the f^-5 and f^-4 terms in the calculation. The zero frequency is a valid input and is important when calculating the surface elevation from the spectrum. We know, however, that the zero frequency should __always__ have zero amplitude as the surface elevation has a mean of zero. This change ensures that if a zero frequency is provided, the amplitude is set to zero.
Previously the surface elevation was only computed using the 'sum of sines' method. This has been found to be prohibitively slow when computing long duration surface elevation traces. This commit introduces the capability to compute the surface elevation using the more computationally efficient IFFT. The IFFT routine is used by default if no frequency bins are provided and the input frequency vector is equally spaced. For example a 1hr sea state realisation at 20Hz took 11s to compute using the 'sum of sines' and 0.007s using the IFFT. Fixes MHKiT-Software#229
|
@mbruggs you da man. Looking this over I think we can get this in our next release for mid-August. I promise to get you a review before then. Currently, I am reviewing several other PRs for the next release so my review of this may be a week or so out. Thank you so much for this contribution! |
|
@akeeste could you review Mark's application of the IFFT? I am looking for help reviewing the application of the technique as I am not familiar with it. |
akeeste
requested changes
Jul 25, 2023
There was a problem hiding this comment.
Hi @mbruggs
Thanks for this great speed update. I only have a few changes to make, which I put in mbruggs#1, including
- add doc string for frequency_bins
- update example notebook calls to surface_elevation to use 0 frequency and the faster IFFT method
Once that is merged, I'll go ahead and bring this into MHKiT
ssolson
added a commit
that referenced
this pull request
Aug 11, 2023
**MHKiT 0.7.0 Release Notes** This release introduces exciting new features and improvements to the MHKiT package: - **Mooring Module**: We are pleased to introduce the new mooring module. This addition primarily supports outputs from MoorDyn. Within this module, users can: - Import data - Calculate lay length - Visualize mooring line movements in 2D and 3D with graphical animations. Accompanying this module is an example notebook to guide users on its functionalities. - **Dolfyn Module Revamp**: The Dolfyn module has been overhauled. Enhancements include: - Turbulence calculation capability - Performance measures for tidal power as outlined in IEC/TS 6200-200. - **New Contributions**: A big shoutout to our community member, @mbruggs, for adding the ability to compute surface elevation using IFFT. - **NDBC Buoy Metadata**: Users can now fetch NDBC buoy metadata directly through MHKiT. - **Delft3D Module Update**: Stay up to date with support for the latest Delft3D NetCDF format. **Additions** - #235 - #232 - #236 - #250 - #239 - #248 **Bug Fixes** - #226 - #238 **Meta/Minor Changes** - #220 - #243 - #225 - #231 - #224 Thank you to all of the contributers who helpped with this release: @mbruggs @Graham-EGI @castillocesar @jmcvey3 @hivanov-nrel @browniea @cmichelenstrofer @akeeste @maxwelllevin @rpauly18 @ssolson
simmsa
added a commit
to simmsa/MHKiT-MATLAB
that referenced
this pull request
Apr 22, 2024
`method` allows the user to choose between using the `ifft` or `sum_of_sines` method for calculation of the surface elevation. See <MHKiT-Software/MHKiT-Python#250> where this was implemented in MHKiT-Python.
This was referenced Apr 22, 2024
simmsa
added a commit
to MHKiT-Software/MHKiT-MATLAB
that referenced
this pull request
May 10, 2024
`method` allows the user to choose between using the `ifft` or `sum_of_sines` method for calculation of the surface elevation. See <MHKiT-Software/MHKiT-Python#250> where this was implemented in MHKiT-Python.
simmsa
added a commit
to simmsa/MHKiT-MATLAB
that referenced
this pull request
May 13, 2024
`method` allows the user to choose between using the `ifft` or `sum_of_sines` method for calculation of the surface elevation. See <MHKiT-Software/MHKiT-Python#250> where this was implemented in MHKiT-Python.
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
As discussed in #229 computing the surface elevation using the 'sum of sines' method is slow. Instead, we can use an IFFT which drastically increases performance. The IFFT method is used as the default when no frquency bins are provided and the frequency vector is equally spaced.
To compute the IFFT the zero frequency must be defined. The spectrum methods have been updated to avoid NaN values in the presence of the zero frequency. This means that spectrum and surface elevations functions work seamlessly together.