search

lanl-ansi / PowerModelsRestoration.jl

A PowerModels Extension for Optimization of Power Network Restoration
https://lanl-ansi.github.io/PowerModelsRestoration.jl/stable/
Other
22 stars 9 forks source link

Fix issues with tests #58

Closed noahrhodes closed 3 years ago

noahrhodes commented 3 years ago

Some MRSP tests do not have consistent solutions because many subsets of items can be selected. Updates remove checking the specific device statuses, and only check that buses are active (to support load) and the objective function.

Removed tests/code for iterative restoration because this algorithm is a work in progress and does not belong in the package yet.

noahrhodes commented 3 years ago

There is one MLD test that results in termination_status=NUMERICAL_ERROR.

result = run_mld(case3_mld_lc, PowerModels.ACPPowerModel, ipopt_solver)

@ccoffrin Do you have an advice for addressing this test?

codecov[bot] commented 3 years ago

Codecov Report

Merging #58 (c92fecd) into master (0ecefa9) will decrease coverage by 1.22%. The diff coverage is n/a.

Impacted file tree graph

@@            Coverage Diff             @@
##           master      #58      +/-   ##
==========================================
- Coverage   92.21%   90.98%   -1.23%     
==========================================
  Files          20       19       -1     
  Lines        1618     1398     -220     
==========================================
- Hits         1492     1272     -220     
  Misses        126      126
Impacted Files Coverage Δ
src/core/constraint.jl 87.05% <0.00%> (-5.89%) :arrow_down:
src/core/constraint_template.jl 92.10% <0.00%> (-3.95%) :arrow_down:
src/form/dcp.jl 92.45% <0.00%> (-1.89%) :arrow_down:
src/prob/rop.jl 96.77% <0.00%> (-1.62%) :arrow_down:
src/core/data.jl 82.67% <0.00%> (-1.09%) :arrow_down:

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 0ecefa9...c92fecd. Read the comment docs.

ccoffrin commented 3 years ago

RE NUMERICAL_ERROR, in this case I usually comment out the test and add a comment note like this, https://github.com/lanl-ansi/PowerModels.jl/blob/460b310288787d4196f9e50b1a81127ee4677a97/test/pf.jl#L19

ccoffrin commented 3 years ago

Seems to pass tests without any issue. What was the NUMERICAL_ERROR bit about?

noahrhodes commented 3 years ago

It gave a numerical error when I tested locally, but interesting that it passed on CI.

ccoffrin commented 3 years ago

Probably there was an update to one of the solvers and you have a slightly older version. In case case, I am going to merge so we can look into the other PR.