Examples¶
The examples are Jupyter notebooks that walk through representative
groundinsight workflows. They are rendered into this site via
mkdocs-jupyter, so all
figures and DataFrame outputs are visible inline.
Notebooks in this section¶
- Minimal example — three-bus MV cable line with a
single source and one fault. The smallest possible working example;
includes the analytical plausibility checks (Kirchhoff at the fault
bus, reduction factor
r = |1 - Z_mutual / Z_self|). Use this as a starting point for your own networks. - MV ring (stationary) — classical frequency-domain fault analysis of a 20 kV ring with 20 substations and a single-core shielded cable type. Shows EPR per bus, reduction factor and grounding impedance at the fault bus.
- MV ring (transient) — same 20 kV ring, full transient stack: FFT vs state-space comparison, the effect of the cable-shield mutual coupling on the fault-bus EPR, and a Thévenin voltage source switched onto the ring.
- Pandapower import — end-to-end use of
gi.preview_pandapower_importandgi.from_pandapower: build a small pandapower MV network, convert it, inject a fault, and run a follow-up outage study on the imported network. - Fault sweep — sweep the fault location across
every bus of a 10-bus MV cable line and visualise the resulting
grounding impedance and reduction factor per fault bus and frequency.
Uses
Network.res_all_impedances()for the bookkeeping. - Fault decomposition — the analysis
layer on top of a solve: the two reduction factors and why they answer
different questions,
analyze_cutsfor what each direction away from the fault contributes,bus_responsefor the response to any electrode at the faulted station from two solves,run_sweepwithsummarize/classifyacross soil resistivities, and the closed-formrun_reference_cases.
Each notebook is self-contained: it builds the network in code, runs the fault calculation, extracts the results as Polars DataFrames and produces the plots inline. Clone the repository to execute them locally: