Commits on Source (38)
-
Thomas Lambert authored
-
Thomas Lambert authored
-
Thomas Lambert authored
-
Thomas Lambert authored
-
Thomas Lambert authored
-
Thomas Lambert authored
-
Thomas Lambert authored
-
Thomas Lambert authored
-
Thomas Lambert authored
-
Thomas Lambert authored
-
Rakotondratsimba Lyraie authored7f58370a
-
Thomas Lambert authored
fix(config): fix caradonna use of airfoil polars See merge request rotare/rotare!1
-
Thomas Lambert authored
-
Thomas Lambert authored
-
204b1295
-
Thomas Lambert authored
Add upstream velocities See merge request rotare/rotare!3
-
Rakotondratsimba Lyraie authored
calculate reynolds number exact value in indvel.m use its exact value when possible, if not use an approximation in ElemPerf.m
1433455a -
Rakotondratsimba Lyraie authored8ed94096
-
Rakotondratsimba Lyraie authored4ad65166
-
Thomas Lambert authored
fix: set true calculation of reynolds number See merge request rotare/rotare!4
-
Rakotondratsimba Lyraie authoredfeb2d069
-
Rakotondratsimba Lyraie authoredebab3f00
-
Rakotondratsimba Lyraie authoredc3a3acbf
-
Thomas Lambert authored
fix(ElemPerf): fix wrong writting of properties See merge request rotare/rotare!5
-
Thomas Lambert authored
This completes the work started in !4 by implementing the true Reynolds recalculation for the other three solvers. Note that special care had to be implemented in Stahlhut solver in order to ensure proper numerical resolution of the solution.
-
Thomas Lambert authored
-
Thomas Lambert authored
-
Thomas Lambert authored
-
Thomas Lambert authored
The functions rotx, roty, rotz are not built-in of MATLAB (wtf). Instead they come from a very expensive toolbox. As none other function of this toolbox are needed, the functions are reimplemented manually for Rotare to remove that useless dependency.
-
Thomas Lambert authored
-
Thomas Lambert authored
-
Thomas Lambert authored
-
Thomas Lambert authored
The old test was sometimes saying that the toolbox was installed when it was not. I do not know why. This seems more robust.
-
Thomas Lambert authored
-
Thomas Lambert authored
Previously, blades of varying airfoils were modeled as multiple consecutive sections with the same pre-definied airfoils. This commit now properly interpolates the geometry, but also most importantly the Polars for intermediate sections between two reference airfoils.
-
Thomas Lambert authored
-
Thomas Lambert authored
The current implementation of the airfoil interpolation leads to a very high computational overhead, as the lift and drag coefficients need to be retrieved for each element one at a time when each element has a slightly different airfoil. The "old way" of doing it (i.e. having the same airfoil until the next reference section), allows for a much faster calculation of the lift and drag coefficients as only the few reference airfoil polars need to be polled for all matching elements at once. If the user inputs enough airfoils, an accurate solution can still be acheived without too much overhead. Simple testing indicate that the complete interpolation leads to a computing time up to 50x larger than without airfoil interpolation.
-
Thomas Lambert authored
The airfoil interpolation process leads to a significant computational hoverhead when the CL and CD coefficients must be calculated for each section, as this process is O(N), N being the number of different airfoils. The basic interpolation as defined previously was creating a specific Airfoil for each section, even when some contiguous sections shared the same profile. This commit tries to limit the total number of Airfoils generated by reusing the same object when it is possible. Currently this only works by trimming airfoils when two contiguous sections share the same profile. Ideally, we could also detect if the same profile appears in non-contiguous sections (i.e. a blade with sections (0012/0015/0012/0012) to reduce it even a bit more. However, as the gain here is expected to be very limited, this final optimization has not been implemented.
Showing
- .gitignore 1 addition, 0 deletions.gitignore
- .gitlab-ci.yml 1 addition, 1 deletion.gitlab-ci.yml
- .gitmodules 3 additions, 0 deletions.gitmodules
- CHANGELOG.md 59 additions, 5 deletionsCHANGELOG.md
- README.md 4 additions, 1 deletionREADME.md
- RELEASE.md 2 additions, 1 deletionRELEASE.md
- miss_hit.cfg 1 addition, 1 deletionmiss_hit.cfg
- src/.miss_hit 6 additions, 0 deletionssrc/.miss_hit
- src/airfoil_data/polargenerator.py 14 additions, 7 deletionssrc/airfoil_data/polargenerator.py
- src/classes/@Blade/Blade.m 38 additions, 20 deletionssrc/classes/@Blade/Blade.m
- src/classes/@ElemPerf/ElemPerf.m 29 additions, 7 deletionssrc/classes/@ElemPerf/ElemPerf.m
- src/classes/@ElemPerf/calcforces.m 1 addition, 1 deletionsrc/classes/@ElemPerf/calcforces.m
- src/classes/@ElemPerf/plotveltriangles.m 3 additions, 3 deletionssrc/classes/@ElemPerf/plotveltriangles.m
- src/classes/@OperRotor/OperRotor.m 1 addition, 1 deletionsrc/classes/@OperRotor/OperRotor.m
- src/classes/@Result/plotperf.m 32 additions, 8 deletionssrc/classes/@Result/plotperf.m
- src/classes/@Rotor/Rotor.m 60 additions, 6 deletionssrc/classes/@Rotor/Rotor.m
- src/configs/caradonna1981.m 10 additions, 2 deletionssrc/configs/caradonna1981.m
- src/configs/knight1937.m 2 additions, 0 deletionssrc/configs/knight1937.m
- src/configs/template.m 3 additions, 1 deletionsrc/configs/template.m
- src/libs/matlab_airfoil_toolbox 1 addition, 1 deletionsrc/libs/matlab_airfoil_toolbox
src/.miss_hit
0 → 100644