Optimization of turboprop model #217
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
Rewrote the formula necessary to compute the turboprop geometry, fuel consumption and maximum thrsut under the form of an OpenMDAO problem. The rationale behind it is that it seems to be quicker to solve it like this which would reduce the computation time of what is right the longest module in FAST-OAD_CS23. Now, what's left to do is implement the use of these components (whose validity were checked in another project) and optimize the implementation, using results caching or a smarter breakdown of the maps
The turboprop sizing has been tested and implemented as a nested problem whose setup is done only once and only when needed
As for the problem, the geometry parameter are computed only when needed
The 4 different problems that will help computing the max thrust at all altitudes have now been implement as was done for the sizing of the turboprop. Note that a problem for the case where propeller thrust is limiting has been added. Their use in the max_thrust function has not yet been implemented !
Reworked the max thrust function so that it uses the nested problem. The main assumption is that, if limits are reached, they are reached in this order : power, opr, itt, propeller thrust. Also added a small cache to the max thrust function for cases like cruise where altitude and speed don't vary
The 2 different problems that will help computing the fuel consumed have now been implement as was done for the sizing of the turboprop. There is a problem with a linesearch algorithm that can be used but will not be favored if possible. There implementation in the sfc function have not yet been implemented
There some differences with the previous version of the code although it is in the order of magnitude of 1-2% for a significant gain in time
Max power is basically computed using the same problem as max thrust is. Except intead of extracting the thrust required we extract shaft power
Implemented a test to check the stauration of the cache
Reworked the cnstruction of the sfc map for the turboprop and checked tested, only minor difference on predicted sfc with new map generation for a considerable speedup (7m to 30s). Now we will test and rework the reading of those maps. Btw `scipy.interpolate.interp2d` is gonna be deprecated so it was changed
Tested the turboprop map construction and read on the TBM900 sizing process. Minimal change on the fuel consumed. Also implemented more widely the use of LinearNDInterpolator to replace interp2d
Codecov ReportAttention: Patch coverage is
❗ Your organization needs to install the Codecov GitHub app to enable full functionality. Additional details and impacted files@@ Coverage Diff @@
## main #217 +/- ##
==========================================
+ Coverage 85.09% 85.61% +0.52%
==========================================
Files 368 396 +28
Lines 26603 28471 +1868
Branches 1447 1464 +17
==========================================
+ Hits 22637 24375 +1738
- Misses 3658 3786 +128
- Partials 308 310 +2 ☔ View full report in Codecov by Sentry. |
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.
The turboprop model was reworked as an OpenMDAO problem rather than relying on scioy to compute fuel consumption and max thrust. Consequently we now have nested problem that are stored as class attributes of the BasicTurbopropEngine. A significant work was done to ensure that these problem are only setup when needed and only once. Also to do subsequent computation we don't "reset" the problem so it is already "warm started". Implemented a pseudo cache for the max thrust function