Skip to content

Built-in plugins catalog

Reference

Prerequisites: What is a plugin

Per-category list of built-in plugins, their params / options, and source location.

Cases (case)

Loaded via pandapower.networks.

name buses source
case9 9 WSCC 9-bus
case14 14 IEEE 14-bus
case30 30 IEEE 30-bus
case39 39 IEEE 39-bus (New England)
case57 57 IEEE 57-bus
case118 118 IEEE 118-bus

Source: pylectra/cases/pp_builtin.py

case_pf: case39

Power-flow (power_flow)

name description options
newton Newton-Raphson (default) tolerance_mva, max_iteration
pandapower pandapower runpp algorithm (nr/bfsw/gs/fdbx/fdxb), tolerance_mva, max_iteration

ODE solvers (ode_solver)

Legacy fixed-step

name algorithm step source
modified_euler Modified Euler case_dyn.stepsize
runge_kutta RK4 same
rkf RK Fehlberg 4(5) same
rkhh RK Higham–Hall 4(5) same

scipy adaptive

pylectra/solvers/scipy_solvers.py

name algorithm when to use
scipy_rk23 RK 2(3) low precision, fast
scipy_rk45 RK 4(5) Dormand–Prince general default
scipy_dop853 RK 8(7) Dormand–Prince high precision
scipy_lsoda LSODA (auto-stiff) ill-conditioned systems
scipy_bdf BDF strongly stiff
scipy_radau Radau IIA 5 strongly stiff + high accuracy

Common options: rtol, atol, max_step, first_step.

torch (pylectra/solvers/torch_solvers.py)

name algorithm adaptive
torch_dopri5 RK 5(4) yes
torch_dopri8 RK 8(7) yes
torch_rk4 RK4 no
torch_euler Euler no

Extra options: chunk_seconds, torch_dtype, device, dense_n. See GPU acceleration.

Faults (fault)

bus_fault — three-phase bus short

fault:
  kind: bus_fault
  params:
    bus: 16             # 1-based
    t_fault: 0.2        # [s]
    duration: 0.05      # [s]

line_trip — line outage

params:
  branch: 21            # 1-based branch row
  t_trip: 0.3
  reclose_after: null   # null = permanent; positive = reclose after that many seconds

load_step — load step change

params:
  bus: 4
  t_step: 1.0
  delta_pd: 100.0       # [MW]
  delta_qd: 30.0        # [MVAr]
  duration: null        # null = permanent

composite — chained events

params:
  events:
    - {kind: bus_fault,  params: {...}}
    - {kind: line_trip,  params: {...}}
    - {kind: load_step,  params: {...}}

Source: pylectra/faults/

Generators (generator)

name order states
classical 2 δ, ω,
two_axis 4 δ, ω, Eq', Ed'

Pgen column conventions in each file's docstring. Source: pylectra/models/generators/.

Exciters (exciter)

name description
simple_avr First-order AVR with cosine voltage feedback
constant Constant Efd (no excitation control)

Governors (governor)

name description states
constant_power dPm/dt = 0 1 (Pm only)
ieee_g 4-state IEEE turbine-governor Pm, P, x, z

PSS (pss)

name description
none No PSS

Scenarios (scenario)

name params
load_perturb sigma_pct (default 5), clip_pct (default 20)
line_outage n_outages (default 1), prob (default 0.5)
noop —

Filters (filter)

name params
pf_converged —
voltage_range vmin (0.85), vmax (1.15), tail_fraction (1.0)
angle_stability max_dev_deg (180)
simulation_completed tol (1e-6)
small_signal_stable margin_max (0.0)

Small-signal (small_signal)

name description
finite_difference Numerical Jacobian via finite differences
modal Same + damping-sorted + eigenvectors by default

Common options: epsilon, method, drop_reference_mode, return_eigenvectors, return_jacobian.

Plots (plot)

name input_kind key kwargs
rotor_angles single relative, gen_indices, palette
speeds single gen_indices
voltages single bus_indices
efds single gen_indices
overview single —
topology case color_by, cmap, bus_size, seed
acceptance batch —
histogram batch column, bins
violin batch column, by
heatmap batch column, rows, cols, aggfunc, cmap

Source: pylectra/plotting/

List registered plugins at runtime

python -m pylectra info

or in Python:

import pylectra
from pylectra import registry
for cat in ["generator", "exciter", "governor", "pss", "ode_solver",
            "power_flow", "fault", "case", "scenario", "filter",
            "small_signal", "plot"]:
    print(cat, registry.list_plugins(cat)[cat])

Next steps