Closed marota closed 4 years ago
I could reproduce your result. Indeed, the score is 0 though the overflow has been resolved The reason is that the total load is greater at timestep 519 than at timestep 518 This rule has been reconducted from Pypownet (alphadeesp.core.grid2op.Grid2opSimulation at line 707)
Problem solved by using:
obs.simulate(action, time_step = 0) in order to avoid changes in loads and productions
This problem is solved and a test should be integrated on this use case to test for topology score @NMegel @mjothy
Done in commit cc033044422aa07134320be30287e2ec8c6ae7e2
using gridPath=yourPath/rte_case14_realistic in config.ini
Running python -m alphaDeesp.main -l 4 -c 0 -t 518
We have a good use case to solve with an overload on line with ID 4
But Topology simulated score are all 0
While many topologies actually solves the problem as can be seen in the plots. For instance topology 112121 at substation with id 1
So Topology simulated score are incorrect