PowSyBl¶

PowSyBl can read the grid and solve the power flow. The optimal power flow always runs in PowerModels, so Julia is still needed.

pip install 'gridfm-datakit[powsybl]'
gridfm_datakit setup_pm

Reading the grid¶

network.reader: powsybl reads the grid with pypowsybl. This is how to load XIIDM, CGMES, PSS/E (.raw), UCTE (.uct) and MATPOWER files.

For a local file, set source: file and put the path to the file, extension included, in file.
For a PGLib case, set source: pglib and put the case name without the pglib_opf_ prefix in name.

Solving the power flow¶

settings.pf_solver: powsybl solves the power flow with Open Load Flow instead of PowerModels. The generator set-points still come from the PowerModels optimal power flow. It requires mode: pf and reader: powsybl, and has no effect in mode: opf.

The default, pf_solver: powermodel, works with both readers.

Generator costs¶

PowSyBl does not read generator costs, including from PGLib. Every generator is given the same cost, c2=0, c1=1, c0=0, so the optimal power flow has no preference for one generator over another.

cost_permutation has no effect, since all costs are identical. cost_perturbation gives each generator its own random cost, so the dispatch varies, but not according to the real costs of the grid. Use none.

Examples¶

Only the keys that differ from scripts/config/default.yaml are shown.

A PGLib case

network:
  name: "case24_ieee_rts"
  source: "pglib"
  reader: "powsybl"

generation_perturbation:
  type: "none"

settings:
  mode: "pf"
  pf_solver: "powsybl"

A local PSS/E file

network:
  name: "my_case"
  source: "file"
  reader: "powsybl"
  file: "grids/my_case.raw"

generation_perturbation:
  type: "none"

settings:
  mode: "pf"
  pf_solver: "powsybl"