Closed MartinBelthle closed 4 months ago
Maybe we should add a small non regression test to be safe
It's actually quite difficult to create such a test case on a 12node network, since it requires the MIP giving different results throughout the iterations, so I guess we should just keep the case that revealed the problem in our sensitive test data.
Please check if the PR fulfills these requirements
Does this PR already have an issue describing the problem? No
What kind of change does this PR introduce? Bug fix
What is the current behavior? In 2nd+ MIP iterations in 2P optimization, we had 0 flow on all preventive cnecs. This caused too-optimistic decisions (0 flow means lot of margin) except when considering loop-flows where we'd have huge loop-flows penalties (0 flow but 1000MW of commercial flow means 1000MW of loop-flows) .
We did so because we disabled the preventive factors to speed-up sensi calculations on curative states without re-enabling it after all calculations were over.
What is the new behavior (if this is a feature change)? We re-enable the factors after the sensi calculations and therefore do not have this problem.
Does this PR introduce a breaking change or deprecate an API?
Other information: Cucumber test are running with this branch to see how many real cases were impacted.