Potential flow wave excitation loads
1 Test description
In this test, we compare wave excitation loads computed by Ashes and OpenFAST according to Potential flow theory. We compare both first order and second order using Newman's approximation.
2 Model
This test uses the Support structure - semisubmersible template. The model in Ashes is rotated 90 degrees so that the surge direction is aligned with the x axis, to match the WAMIT coordinate system. The .3 and ,12d files used to compute the hydrodynamic loads are taken from the OpenFAST regression test repository, here: https://github.com/OpenFAST/r-test/tree/main/glue-codes/openfast/5MW_Baseline/HydroData.
The model is constrained so there are no radiation damping or added mass forces.
The load cases tested are decribed in the following table:
| Load case | Wave angle (Ashes) | WaveDir (FAST) | Regular/Irregular | H/Hs | T/Tp | 2nd order loads computed? |
| LC1 | -90 | 0 | Regular | 5 | 9.66 | Yes |
| LC2 | -90 | 0 | Irregular | 6 | 10 | Yes |
| LC3 | -110 | 20 | Regular | 5 | 9.66 | No |
| LC4 | -70 | -20 | Regular | 5 | 9.66 | No |
| LC5 | -110 | 20 | Irregular | 6 | 10 | No |
| LC6 | -70 | -20 | Irregular | 6 | 10 | No |
In OpenFAST, the wave direction is defined by the parameter WaveDir in the SeaState input file.
In Ashes, the default wave direction is from the positive y-axis toward the negative y-axis (i.e., from North). To obtain waves traveling from the positive x-axis toward the negative x-axis in Ashes (corresponding to WaveDir = 0 in OpenFAST), a wave direction of -90° must be specified.
Furthermore, a positive wave direction in Ashes rotates the wave propagation towards the East (from positive x toward negative x), whereas a positive WaveDir in OpenFAST rotates the waves counterclockwise, i.e. from negative y toward positive y.
This difference in conventions—admittedly confusing—is illustrated in the figure below:
At the time of writing this test, the OpenFAST regression repository does not provide a .12d file with non-zero wave direction. Therefore, a different .12d file (which does not correspond to the DeepCWind semisub) was used. However, OpenFAST could not import that file because of the frequency discretisation. Therefore, 2nd order loads are not computed for the Load cases with a non-zero wave direction.
3 Benchmarked solutions
The first and second order potential flow wave excitation loads were produced with OpenFAST 4. All degrees of freedom are disabled. For irregular waves, the wave elevation is produced by Ashes and read by OpenFAST.
4 Results
A simulation of 60 seconds is run for each load case. The test is considered passed if the results from Ashes are within 0.1% of those from OpenFast.
The report for this test can be found on the following link: