pandapower-sim

name: pandapower-sim description: Use when driving pandapower (Python power-system analysis library combining pandas data tables with PYPOWER-style solvers, from Fraunhofer IEE) via Python scripts — load flow, optimal power flow (OPF), short-circuit, time-series, contingency analysis on transmission / distribution networks — through sim runtime one-shot execution.

You are connected to pandapower via sim-cli.

pandapower (Thurner et al., IEEE Trans. Power Systems 2018) is the standard Python library for power-system analysis. Pure Python + pandas + numpy/scipy (with optional Numba for speedup). Pip-installable (pip install pandapower).

Scripts are plain .py:

import pandapower as pp
net = pp.create_empty_network()
b1 = pp.create_bus(net, vn_kv=20.0); b2 = pp.create_bus(net, vn_kv=20.0)
pp.create_ext_grid(net, b1)
pp.create_line(net, b1, b2, length_km=10, std_type='NAYY 4x150 SE')
pp.create_load(net, b2, p_mw=0.5)
pp.runpp(net)
print(net.res_bus.vm_pu)            # bus voltage magnitudes (p.u.)

Same subprocess driver mode as PyBaMM.

base/ — always relevant

Path	What's there
`base/reference/workflow.md`	Empty net → buses → branches → injections → run.
`base/reference/elements.md`	Bus / Line / Trafo / Load / Gen / SGen / ExtGrid / Storage / Switch.
`base/reference/solvers.md`	runpp (PF), runopp (OPF), runsc (short-circuit), runts (time-series).
`base/reference/standard_libs.md`	Built-in line types, transformer types, IEEE benchmark networks.
`base/snippets/01_pf.py`	Verified 2-bus power flow E2E.
`base/known_issues.md`	numpy bool serialization, AC vs DC PF, slack bus required.

sdk/2/ — pandapower 2.x

sdk/2/notes.md — version-specific surface notes.

Hard constraints

Every network needs an ext_grid (slack bus) for AC power flow, else runpp raises. For islanded systems use a gen with slack=True.
Bus voltages set by vn_kv — must match the line / transformer ratings on each side. Mismatched voltages silently cause convergence failures.
Default runpp is AC Newton-Raphson. For approximate / DC use runpp(net, algorithm='dc'). For three-phase: runpp_3ph.
Acceptance != "ran without error". Always check that PF converged and bus voltages are within plausible bounds (typically 0.95-1.05 p.u.) and line loadings < 100%.
Result tables live on net.res_* AFTER running. Empty before.
Print results as JSON on the last stdout line. Cast pandas floats: float(net.res_bus.vm_pu[b]).

Required protocol

Gather inputs:
- Category A: network topology (bus / line / trafo data), load and generation profiles, slack reference, acceptance.
- Category B: PF algorithm, tolerance, max iterations.
sim check pandapower.
Write .py per base/reference/workflow.md.
sim lint script.py.
sim run script.py --solver pandapower.
Validate JSON.