OpenFAST v3.2.0 #1175
OpenFAST v3.2.0 #1175
Conversation
Does not run (seg faults) since there are old routines still that need updating.
- new order: Blade, nacelle, tower, substructure - The order is very important in the linearization, so I reordered simply to keep it consistent everywhere.
- Update SrvD_JacobianPInput - still missing the dYdu user channels - also missing dXdu - Update GetOP for dx_op matrix - Updates to FAST_Lin.f90 so it now compiles and actually runs!!!! Just produces rubbish... There was some confusion on a prior commit about where the logmapping flag from FAST_Lin goes (doesn't go with the u% stuff, only for y% motion DCMs)
- this is necessary so I can set outputs for single StC during linearization (speeds up code)
- think this is working for BStC
- but u (unerturbed) was getting used to set the StC motion input perturbations :(
Increase rank of arrays so that blade number and StC instance are tracked. This simplifies assembling of jacobians in glue code.
Also a few blank lines moved for easier git comparisons
fix a few typos in Linear_SrvD_InputSolve_du, and add some comments
* MAP: Jacobian_PInput modifies y, but should be at y_op (See #1165) * MAP: dll routine to return summed load at operating point
Ok done! |
|
I'd suggest also adding this bug fix: #1158 |
- Changed name of optional parameter in`ModName_GetOP` routine from `NeedLogMap` to `NeedPackedOrient` in an attempt to clarify what that does (returns the orientation matrix as a 3-valued vector instead of 9-valued one) - Made the use of this value consistent in ElastoDyn (all meshes with orientation fields that are packed in the GetOP routine need to have the same value for their call to `PackMotionMesh`) - Added the optional `NeedPackedOrient` input to `SD_GetOP` - Made ED, BD, and SD each pack their arrays the same way in `FAST_DiffInterpOutputs`; otherwise, the linear trim solution does not difference all of the outputs correctly to determine if it has converged
Thanks - did it! |
There was a problem hiding this comment.
I'm happy to see OpenFAST v3.2.0 finally being released! Please find a few comments regarding the updated documentation below.
Also, after finally reviewing #1050 that has already been merged, I found a couple discrepancies wrt the 3D rotation implementation plan. I've now sent these comments to @andrew-platt and will paste them here for completeness. These changes should also be implemented in this release:
-
For the perturbation of input rotations for the TStc and SStc meshes in ServoDyn, the small angle perturbation approach should be used, i.e., the call to PerturbOrientationMatrix() needs UseSmlAngle=.true. For BStc and NStc, the implementation looks correct.
-
Likewise in AeroDyn for the tower motion orientation, UseSmlAng = .true. should be used.
…cture (SStc) This now aligns with the 3D rotation implementation plan (in docs/OtherSupporting).
Corrections for 3.2.0 release (jjonkman review)
Feature or improvement description
Pull request to merge
mainintodevand create a tagged release forv3.2.0.See the milestone and project pages for additional information
Test results, if applicable
See GitHub Actions
Release Checklist
Proposed Change Log
General
#866 Include legacy documentation in pdf and MS Word format - Modules
#1021 Reg_tests python scripts: fix issue where directory returned is empty
#1062 Add documentation for community contribution
#1169 Update git-module urls
#1158 Update orientation differences in linear trim solution
#1177 Add documentation on 3D rotations in linearization
#1178 Corrections for 3.2.0 release
AeroDyn
#932 BugFix: UA update states that were not updated
#1032 AD: added more nodal outputs for debugging UA
#1039 [BugFix] Nacelle motion not passed between AeroDyn driver and AeroDyn
#1043 AeroDyn: combine some FVW and BEMT output calculations
#1045 OLAF: check for division by zero to avoid invalid calculations
#1141 Move AD module reg tests to a standalone job
BeamDyn
#1050 Fix aeroelastic stability analysis with BeamDyn
FAST.Farm
#1107 Fix Time-Step Delay in Super Controller within FAST.Farm
HydroDyn
#806 C-bindings interface for HydroDyn
#1047 Fix HydroDyn Jacobian outputs when LinOutJac is True
#1108 BugFix: Checks to populate matrix input to Newman's app were reversed
#1173 Bug Fix: rotation matrix perturbation with small angles was wrong
InflowWind
#1144 BugFix: Fix typo in DLLEXPORT attributes for IfW C interface
MAP++
#1048 MAP: small fix based on MAP 1.3
#1168 Fix MAP linearization operating point
NWTC Library
#1050 Fix aeroelastic stability analysis with BeamDyn -- changes to the library routines for angle perturbations during linearization
#1124 Updates to VersionInfo module and its dependencies
#1157 Fix for binary file compression
OpenFAST Library
#962 Add a Python glue-code interface
#1057 FAST Library: add access to hub position and velocity
ServoDyn
#803 Linearization of ServoDyn Structural control elements
#1074 [BugFix] ServoDyn StC control signal channels were not zeroed properly
#1089 BugFix: Fix API change line numbers
#1101 BugFix: ServoDyn API change docs
#1140 Add non-rotating hub forces to Bladed Interface
#1160 Correction to documentation on Extended Bladed DLL interface
TurbSim
#887 TurbSim modifications