Transmission Network Analysis

A Python toolkit for aggregating and visualizing electricity transmission networks at the county level from MATPOWER-style grid datasets. Built using the comprehensive USA Test System dataset (Xu et al., 2021) from Zenodo.

🔧 Features

• Transmission Aggregation: Convert MATPOWER bus/branch data to county-level transmission edges
• HVDC Support: Analyze both AC and HVDC transmission lines
• Geospatial Mapping: Create detailed maps with capacity-based line visualization
• Interactive Visualizations: Generate interactive HTML maps with hover details
• Regional Filtering: Focus on specific states, zones, or interconnects
• Capacity Classes: Visual distinction across 6 capacity ranges (<200MW to 5K+MW)

📊 Dataset Overview

The processed dataset contains 7,652 transmission edges with:

• 7,637 AC transmission lines
• 15 HVDC transmission lines
• 21 data columns including geographic, network, and capacity information
• Complete US coverage across Eastern, Western, and Texas interconnects

🚀 Quick Start

1. Installation

git clone https://github.com/swang22/Transmission-Network-Aggregator.git
cd Transmission-Network-Aggregator
pip install -r requirements.txt

2. Data Preparation

This toolkit uses the USA Test System dataset from Zenodo:

Xu, Yixing, et al. "US test system with high spatial and temporal resolution for renewable integration studies." 2020 IEEE Power & Energy Society General Meeting (PESGM). IEEE, 2020. A Synthetic Time-Variant U.S. Power Grid Data Set for the Evaluation of Variable Generation and Demand Response. Zenodo. https://doi.org/10.5281/zenodo.4538590

The dataset provides a comprehensive synthetic representation of the U.S. power grid with time-variant generation and demand data. For transmission analysis, we use the network topology files:

# The base_grid folder should contain:
# - bus.csv, branch.csv, sub.csv, bus2sub.csv
# - dcline.csv (for HVDC), zone.csv

Dataset Features:

• 82,000+ buses representing generation and load points
• 104,000+ transmission branches with detailed electrical parameters
• Geographic mapping to U.S. counties via substations
• Three interconnects: Eastern, Western, and Texas (ERCOT)
• Voltage levels: From distribution (4kV) to transmission (765kV)

To download the original dataset:

1. Visit: https://zenodo.org/records/4538590
2. Download the USATestSystem.zip file (~4.2 GB)
3. Extract the network data files to the data/base_grid/ folder

Note: The repository includes a cleaned subset focused on transmission analysis (49 MB vs 4.2 GB).

3. Generate Transmission Data

python src/run_transmission.py

This creates outputs/county_edges_tx.csv with all transmission edges.

4. Create Visualizations

# Texas transmission network
python src/visualize_transmission.py --region "TX" --type state --output outputs/texas.png

# Interactive California map
python src/visualize_transmission.py --region "CA" --type state --output outputs/california.html --interactive

# Western Interconnect overview
python src/visualize_transmission.py --region "Western" --type interconnect --output outputs/western.png

Advanced Filtering Options

Voltage-Based Filtering

# Transmission-only analysis (≥138 kV) - RECOMMENDED
python src/run_transmission.py --transmission-only

# Custom voltage threshold
python src/run_transmission.py --min-voltage 230

# Specific voltage levels only
python src/run_transmission.py --voltage-levels 345 500 765

# Compare filtering impact
python src/run_transmission.py --output outputs/all_voltages.csv
python src/run_transmission.py --transmission-only --output outputs/transmission_only.csv

Analyze Your Dataset

# Understand voltage levels in your data
python examples/voltage_analysis.py

# Shows impact: transmission filtering removes ~33% of edges (distribution connections)
# Result: 7,652 edges → 5,157 edges (cleaner bulk power transfer analysis)

5. Run Examples

python examples/generate_examples.py

⚡ Voltage Filtering Impact

The --transmission-only filter provides significant improvements for power systems analysis:

Metric	All Voltages	Transmission (≥138 kV)	Improvement
Branches processed	81,861	41,048	50% reduction
County-to-county edges	7,652	5,157	33% cleaner
Focus	Mixed levels	Bulk power transfer	More accurate

Why filter by voltage?

• Prevents capacity inflation from local distribution connections
• Focuses on true transmission corridors between regions
• Follows standard power systems definitions (138+ kV = transmission)
• Eliminates noise from sub-transmission interconnections

🗺️ Visualization Examples

Texas Transmission Network

Capacity Classes

The visualization system uses 6 capacity classes for clear visual distinction:

• < 200 MW: Thin lines (local distribution)
• 200-500 MW: Light lines (sub-transmission)
• 500-1K MW: Medium lines (transmission)
• 1K-2K MW: Thick lines (high voltage)
• 2K-5K MW: Very thick lines (extra high voltage)
• 5K+ MW: Thickest lines (ultra high voltage corridors)

📁 Project Structure

transmission-network-analysis/
├── src/                          # Core source code
│   ├── run_transmission.py       # Main aggregation script
│   ├── visualize_transmission.py # Visualization engine
│   └── grid2county_txcap.py     # Aggregation functions
├── examples/                     # Example scripts and outputs
│   ├── generate_examples.py     # Create sample visualizations
│   └── README.md                # Examples documentation
├── data/                        # Input data
│   ├── base_grid/              # MATPOWER grid data
│   └── counties/               # US county shapefiles
├── outputs/                     # Generated results
│   ├── county_edges_tx.csv     # Main output dataset
│   └── README.md               # Output documentation
├── tests/                       # Unit tests
├── docs/                        # Documentation
├── requirements.txt            # Python dependencies
├── .gitignore                 # Git ignore rules
└── README.md                  # This file

🔍 Data Schema

The output county_edges_tx.csv contains 21 columns:

Geographic Information

• from_fips, to_fips: County FIPS codes
• from_county, to_county: County names
• from_state, to_state: State abbreviations
• from_lat, from_lon, to_lat, to_lon: Coordinates

Network Information

• from_zone_id, to_zone_id: Zone identifiers
• from_zone_name, to_zone_name: Zone names
• from_interconnect, to_interconnect: Interconnect regions

Transmission Details

• total_capacity_mw: Combined transmission capacity
• edge_type: 'AC' or 'HVDC'
• num_circuits, num_links: Circuit/link counts
• total_impedance: Electrical characteristics

🎯 Use Cases

• Grid Planning: Identify transmission bottlenecks and expansion needs
• Policy Analysis: Analyze inter-regional transmission capabilities
• Research: Academic studies on power system topology
• Education: Teaching power systems geography and capacity
• Visualization: Create publication-quality transmission maps

🧪 Testing

Run the test suite:

python -m pytest tests/

📖 Documentation

Detailed documentation is available in the docs/ folder:

• API Reference
• Data Format Guide
• Visualization Guide

🤝 Contributing

Contributions are welcome! Please see our Contributing Guide for details.

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

Original Dataset

This work uses the USA Test System dataset:

• https://doi.org/10.5281/zenodo.4538590

Software & Libraries

• MATPOWER for the grid data format specification
• US Census Bureau for county boundary shapefiles
• Python Geospatial Community (GeoPandas, Shapely, PyProj) for spatial analysis tools
• Visualization Libraries (Matplotlib, Plotly) for mapping capabilities

Research Context

The original dataset was developed to support research in:

• Variable renewable energy integration
• Demand response optimization
• Power system planning and operations
• Grid resilience and reliability analysis
• US Census Bureau for county shapefiles
• Python geospatial community (GeoPandas, Shapely, etc.)

---

For questions about the underlying grid dataset or methodology, please refer to the documentation or open an issue.