DC line implementation #143
Conversation
…ine remaining, but the solution looks good.
…, updated generator cost function reading
| PetscInt Nbus, Ngen, Nline, Nload, | ||
| Ndcline; /* global # of buses,gens,branches,loads, and dclines (includes | ||
| elements which are out of service */ | ||
| PetscInt nbus, ngen, nline, nload, | ||
| ndcline; /* local # of buses,gens,branches,and loads */ | ||
| PetscInt NlineON, nlineON; /* Number of active lines (includes DC lines) */ | ||
| PetscInt NdclineON, ndclineON; /* Number of active dc lines */ | ||
| PetscInt NgenON, ngenON; /* Number of active generators */ | ||
| PetscInt Nref, nref; /* Number of reference buses */ | ||
| /* Number of generator types */ | ||
| PetscInt ngencoal, ngenwind, ngensolar, ngenng, ngennuclear, ngenhydro, | ||
| ngenundefined; | ||
| /* Number of renewable generators (solar, wind) */ | ||
| PetscInt ngenrenew; | ||
| /* Number of isolated buses */ | ||
| PetscInt nisolated_buses, Nisolated_buses; | ||
| PetscInt *isolated_buses; /* Array to hold isolated buses */ |
There was a problem hiding this comment.
I am not thrilled about a lot of these variable names. Is there a better way to differentiate / describe these variables? The difference of 1 letter in a different case is easy to typo / miss, and the intention isn't clear from reading.
There was a problem hiding this comment.
N and n are typically used in the context of a parallel implementation where N is the global size and n is the local size. Perhaps, it is confusing here and this should be replaced. But, this is a big change that we need to do through another PR.
There was a problem hiding this comment.
Should we create an issue for this then? Perhaps naming n_local_... and n_global_... would make more sense? A little more of a cumbersome name, but should be more readable.
|
@cameronrutherford : I've addressed your comments. Please take a look and let me know if anything else is needed. Thanks. |
Lot's of CI failures on both deception and Newell. Hard for me to say what the fix is, but can't really merge this until tests are passing. Would love to see some documentation as well. |
|
@cameronrutherford : I have reverted the angle limits back to negative and positive infinity. Hopefully, this should fix the tests. |
|
@abhyshr still lots of errors in CI tests We can easily update the tests to respect the new limit if you want to revert back to your original change, but these Ipopt based errors seem to be independent from your change |
Merge request type
Relates to
This MR updates
Summary
DC line implementation for PFLOW and OPFLOW
Summarize the MR concisely
This PR adds DC line model in PFLOW and OPFLOW. An input file for testing
datafiles/case9/case9_dcline.mhas been also added along with a toml test and a smoke test.