jordan-model — An Open Synthetic Test System for the Jordanian Transmission Grid

Repository for the paper "An Open Synthetic Test System for the Jordanian Transmission Grid" (submitted, IEEE Transactions on Power Systems). Calibration year 2018. Every number in the paper regenerates from the public inputs and scripts here.

Preliminaries

A synthetic, validated transmission test system for the Jordanian national grid — 81 solved nodes (68 substations, 13 split 400/132 kV), a plant-level 2018 thermal and renewable fleet with per-site hourly profiles, calibrated hourly load, and a full validation suite. The topology is reconstructed from a published PMU-placement diagram and restored to its 2018-energized state using dated NEPCO records. This is a synthetic test case, not the operational NEPCO model; see the paper's scope statement before use.

Layout

data/      inputs (hand-built / extracted): topology edges, RE fleet,
           bus-substation identity key, coordinates, named line lengths
src/       the pipeline, in run order: profiles, bias derivation, load,
           annual PCM, network build + N-1, structural metrics, losses,
           ex-post congestion check, cross-solver, scenario, figures
outputs/   released artifacts: jordan_net.json, jordan_net_greencorridor.json,
           case_jordan.mat (corrected MATPOWER export), generated hourly
           series (shipped so no API keys are needed), dispatch and
           energy-mix tables, named N-1 contingency list
archive/   Development versions - kept for history, do NOT
           reproduce the paper

Reproduction

bash run_pipeline.sh                       # profiles -> load -> PCM -> network -> N-1 -> metrics
python src/loss_decomposition.py outputs/jordan_net.json outputs/load_2018_hourly.csv outputs/annual_dispatch.csv
python src/cross_solve.py export           # + MATLAB step (see docstring), then: compare
python src/check_congestion.py             # ex-post hourly network feasibility
python src/scenario_green_corridor.py

Regenerating the weather-driven profiles from scratch is optional (the generated CSVs are in the repo) and requires an NSRDB API key and a Renewables.ninja token in src/re_profiles_2018.py.

Validation protocol

Parameters are set from cited sources or labeled assumptions before any comparison with the 2018 record. Algorithm-correctness fixes are permitted with a single rerun and then frozen. No parameter is ever adjusted against the validation year. Two disclosed calibrations exist (the Risha fuel-deliverability cap and the load temperature coefficient), both to calibration anchors, never to validation targets.

Requirements

Python 3.9+, pandapower, pandas, numpy, networkx, matplotlib, pvlib, scipy. The cross-solver step additionally needs MATLAB or Octave with MATPOWER.

License and citation

Code is released under the MIT License; the data files under CC BY 4.0. If you use this test system, please cite the paper (citation to be updated on publication).