PowerGridPlanning.jl

A Julia package for transmission grid planning on realistic power system networks. Provides DC and linearized AC formulations for co-optimizing line switching, line hardening, battery storage siting, and solar PV siting against load shedding, cost, and risk-exposure objectives, plus nonlinear AC verification/recovery runs for checking planned decisions. Applications include planning under severe-weather risk such as wildfires, with built-in support for USGS Fire Potential Index data.

Overview

This package provides a unified interface for transmission grid planning problems on realistic power system networks. Operational decisions (line switching) and capital investments (line hardening, battery storage siting, solar PV siting) are co-optimized under a single objective — load shedding, generation cost, risk exposure, or a weighted tradeoff. The per-line risk interface is hazard-agnostic; the package ships with built-in data loaders for severe-weather risk applications, currently wildfire risk via the USGS Fire Potential Index.

Key Features:

• Four formulations: wildfire-aware switching (DCOTS, LACOTS) and pure power-flow baselines (DCOPF, LACOPF) sharing the same investment-planning interface
• Nonlinear AC verification with diagnostics: Replay fixed planning decisions with AC power flow (ACPF) or run AC redispatch/recovery with load shedding (ACOPF), including voltage, thermal, islanding, and recovery-stress diagnostics
• Multiple objective functions: Load shedding minimization, risk-exposure minimization, generation cost minimization, and customizable tradeoffs
• Two switching methods: Optimal MIP-based and fast thresholded heuristic
• Line hardening: Optimize infrastructure investments (vegetation management, covered conductors, or undergrounding) to permanently reduce per-line risk exposure
• Battery energy storage systems (BESS): Optimize battery installation and operation for load shedding mitigation
• Solar PV installation: Optimize solar capacity placement with hourly capacity factors and inverter reactive power support (LACOTS)
• Hazard-agnostic risk interface: Accept any per-line risk signal; built-in loader for USGS Fire Potential Index (wildfire) included
• Multi-period optimization: Solve for single days, specific date ranges, or entire months/years
• Flexible network support: Pre-configured for 6+ realistic power system test cases (RTS-GMLC, CATS, Texas7k, ACTIVSg2000/10k, WECC240), or bring any MATPOWER case via :case_file (see the Usage Guide)

Installation

Prerequisites

• Julia 1.10 or higher (LTS; 1.6+ is technically compatible but 1.10 is recommended)
• Gurobi optimizer with valid license for planning solves (solve_ots; academic licenses available)
• Ipopt is included as the default nonlinear solver for AC verification (verify_ac)

Setup

Option A — install from the Julia General registry (recommended):

using Pkg
Pkg.add("PowerGridPlanning")
using PowerGridPlanning

To track the latest development version on main instead of the registered release:

Pkg.add(url="https://github.com/rpiansky3/PowerGridPlanning.jl")

Option B — clone and develop locally:

1. Clone this repository:

git clone https://github.com/rpiansky3/PowerGridPlanning.jl.git
cd PowerGridPlanning.jl

1. Instantiate the package environment:

julia --project=. -e 'using Pkg; Pkg.instantiate()'

1. Test the installation:

julia --project=. -e 'using PowerGridPlanning; println("OK")'

Quick Start

The repository ships with a reference dataset in test_data/ covering June 2020 for all six networks. Use :data_dir => "test_data" to run immediately after cloning — no additional data download required.

using PowerGridPlanning

# Works out of the box using the included test_data/ reference dataset
opt_parameters = Dict(
    :network   => "RTS",
    :model     => "DCOTS",
    :objective => "loadshed",
    :times     => [(2020, 6, 15)],   # June 15, 2020 — available for all 6 networks
    :data_dir  => "test_data"
)

results = solve_ots(opt_parameters)

println("Solve time: $(results[:solve_time]) seconds")
println("Total load shed: $(results[:total_load_shed]) MW")
println("Risk reduction: $(results[:risk_reduction_pct])%")
println("Lines switched off: $(length(results[:switched_off_lines][1]))")

To verify all six networks load and solve correctly:

julia --project=. scripts/verify_reference_data.jl

Tutorial

A Jupyter notebook walkthrough is included at tutorial.ipynb. It covers the core API end-to-end on the RTS network — basic solve, switching methods, hardening, battery and solar siting, nonlinear AC verification/recovery, the tradeoff curve, and plotting — using only the bundled test_data/ so it runs out of the box after Pkg.instantiate() and a Gurobi license.

julia --project=. -e 'using IJulia; notebook(dir=".")'

Then open tutorial.ipynb from the Jupyter file browser.

Where to go next

• Models and Methods — the DCOTS/LACOTS/DCOPF/LACOPF formulations, switching methods, objectives, and hardening
• Data — bundled reference data, supported networks, and the solar/census/wildfire data pipelines
• Usage Guide — every solve_ots, solve_opf, and verify_ac parameter
• Command-Line Interface — running studies from the terminal
• Examples — 17 worked examples
• Results Dictionary — every key in the results
• Plotting — geographic overviews, dispatch time series, and tradeoff curves
• API Reference — docstrings for the public API

Citation

If you use this package in your research, please cite:

@software{PowerGridPlanning_jl,
  author = {Piansky, Ryan},
  title = {PowerGridPlanning.jl: Transmission Grid Planning with Co-Optimized Switching, Hardening, and Storage},
  year = {2026},
  version = {0.2.0},
  url = {https://github.com/rpiansky3/PowerGridPlanning.jl}
}

License

This package is released under the MIT License.