Merge rc-3.5.1 into dev #1709
Merged
Merge rc-3.5.1 into dev #1709
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
…date Update WAMIT .hst input file and baselines for floating MHK r-test
…date Update HydroDyn input file and baselines for floating MHK r-test
The input arrays for Lidar to ServoDyn get allocated in ServoDyn.f90, and again in BladedInterface_EX.f90. This double allocation causes issues for many compilers. The logic behind all this really should be looked at in more detail, but since the Lidar module in InflowWind will be superceeded by LidarSim in release 4.0.0, I'm not going to sort the logic right now.
Affected arrays:
u%LidSpeed
u%MsrPositionsX
u%MsrPositionsY
u%MsrPositionsZ
Thanks to @foxton9 for reporting the issue, and proposing the fix!
This was reported in issue OpenFAST#1484
[BugFix] ServoDyn inputs for Lidar allocated twice, and a few other minor issues
[BugFix] HD wave visualization with 2nd order waves
The last revision had the incorrect case for the CMake variable, so the flag wasn't being added
We added the hub point to AWAE at the center of the grid now that IfW calculates the disk average velocity every time IfW_CalcOutput gets called. But this point needs to be shifted by the turbine position so it is in the box. This will all change in 4.0.0 when we switch to pointers for all this and don't call IfW_CalcOutput directly anymore.
Fix adding OpenMP flags to CMAKE_Fortran_FLAGS
FF: fix bug with Mod_AmbWind=3 -- hub point wasn't shifted.
This is a hack fix -- the logic really should be improved elsewhere in the code for checking if any blades are output. However, we don't actually have logic in place yet for only outputting certain blades.
This was causing issues in the linearization as the reported windspeed was on the order of 1E+38, which did not format correctly in the `.lin` output file
ED/AD: fix segfault when BldNd_BladesOut==0
[BuFix] InflowWind was not setting InitOut MWS data.
AD: Adding rtest for TailFin polar-based model
When building the Simulink MEX file with CMake on Linux, the resulting shared library had unresolved symbols. This was due to the order in which the static libraries were specified in simulink/CMakeLists.txt. The GCC linker does one pass over the libraries and notes unresolved dependencies which can only be resolved by the succeeding libraries. This means that the most generic libs (like NWTC-Library) need to appear at the end of the list. Instead of leaving it up to the user to get the order right, the simulink/CMakeLists.txt file has been updated to link the libraries twice. The same issue occured when using USE_LOCAL_STATIC_LAPACK in that the BLAS libraries were linked before LAPACK so there were unresolved symbols. This has been fixed by reversing the order. Also, the OpenFOAM libraries have been marked as STATIC.
* AD: more bldnd outputs channels, more documentation of future bem * AD: renaming psi as psi_s * AD: mode outputs channels * AD: more comments in BEM code * AD: documenting coordinate systems * AD: implementing additional outputs for Buoyancy * AD: using i instead of g in AeroDyn regular outputs * AD: wrapped optional argument call to avoid passing them if not provided * AD: update of coordinate system documentation * AD: change comments polar (p) -> local-polar (l) * AD: links were swapped in documentation
* RTest: update of lin file comparison * Update of lin baselines * RTest: adding freq and damp comparisons for lin tests * RTest: BD_Free_Free case is difficult for damping comparison
Remove deleted variables BlFB and BlMB from AeroDyn_AllBldNdOuts_IO
ebranlard
approved these changes
Jul 28, 2023
|
I suspect that because of #1694 we might have to update the baselines of the lin cases of dev a bit more often until we find a better solution. We might need to introduce some test specific "if" statements in the python code. It seems that sometimes negative frequencies and highly damped modes show up in the eigenvalue analysis. We could try and increase the output fmt resolution of the lin files, that should help the robustness of the EVA. For now I would not worry if only the lin cases are failing... |
Update r-tests
Disabling this test case for now so PR OpenFAST#1709 can be completed as it is holding up development. The test passes locally, but takes >1500 sec to run on GH Actions which causes it to fail with a timeout. A task has been created in GH so this gets fixed before 4.0.0 is released.
See if 5MW_OC4Semi_Linear will run faster when built using OpenBLAS
andrew-platt
approved these changes
Jul 31, 2023
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.
This PR is ready to be merged.
Feature or improvement description
This PR merges the
rc-3.5.1branch into thedevbranch to keep the developments in sync. It also changes the Github Actions to use LAPACK provided by OpenBLAS which significantly reduces the time to run the linearization regression tests.Related issue, if one exists
#1688, #1703, #1704, #1705
Impacted areas of the software
HydroDyn, SeaState, AeroDyn, Regression Tests
Test results, if applicable
The baselines for
5MW_Land_BD_Linear,Fake5MW_AeroLin_B1_UA4_DBEMT3, andFake5MW_AeroLin_B3_UA6were updated. All other regression tests were unchanged after resolving merge conflicts.