Installation#

This guide covers the complete installation of PyPSA-GB.

Step 1: Install Conda#

If you don’t have Conda installed, download and install Miniconda or Anaconda.

Verify your installation:

conda --version

Step 2: Clone the Repository#

Clone PyPSA-GB from GitHub:

git clone https://github.com/andrewlyden/PyPSA-GB.git
cd PyPSA-GB

Note

The repository is approximately 2GB due to included data files. This may take some time depending on your internet connection.

Step 3: Create the Conda Environment#

Create the pypsa-gb environment using the provided environment file:

conda env create -f envs/pypsa-gb.yaml

Activate the environment:

conda activate pypsa-gb

Tip

Add conda activate pypsa-gb to your shell profile to automatically activate it in new terminals.

Step 4: Install a Solver#

PyPSA-GB requires an optimization solver. You have two options:

Option A: Gurobi (Recommended)#

Gurobi offers free academic licenses and provides excellent performance.

Register for an academic license
Download and install Gurobi
Activate your license:
```
grbgetkey YOUR-LICENSE-KEY
```

Option B: HiGHS (Open Source)#

HiGHS is included in the conda environment. No additional setup required.

To use HiGHS, set in your scenario configuration:

solver:
  name: "highs"

Step 5: Verify Installation#

Test that everything is working:

# Check Python and key packages
python -c "import pypsa; print(f'PyPSA version: {pypsa.__version__}')"
python -c "import snakemake; print('Snakemake OK')"

# Dry run to check workflow
snakemake -n -p

Step 6: Download Weather Data#

For renewable generation profiles, you need ERA5 weather data “cutouts”. PyPSA-GB uses a tiered acquisition strategy to minimize download times:

Data directory - Check data/atlite/cutouts/ for cached copy (instant)
Zenodo - Download pre-built cutouts from Zenodo repository (~5-10 minutes per year, years 2010-2024)
ERA5 API - Full download via atlite as fallback (~2-4 hours per year)

Quick Start (Recommended)#

For years 2010-2024, cutouts are automatically downloaded from Zenodo:

# Edit the configuration to specify which years you need
nano config/cutouts_config.yaml

# Set years_to_generate, for example:
# years_to_generate:
#   - 2020
#   - 2021

# Generate cutouts (will download from Zenodo for 2010-2024)
snakemake -s Snakefile_cutouts --cores 1

No CDS API credentials required for years 2010-2024 when using Zenodo!

Manual ERA5 Download (Advanced)#

For years outside 2010-2024 or if you prefer direct ERA5 download:

Configure CDS API (one-time setup):

# Register at: [https://cds.climate.copernicus.eu/](https://cds.climate.copernicus.eu/)
# Get your API key from: [https://cds.climate.copernicus.eu/how-to-api](https://cds.climate.copernicus.eu/how-to-api)
# Create ~/.cdsapirc with your credentials

Disable Zenodo (optional) in config/cutouts_config.yaml:

zenodo:
  enabled: false  # Force ERA5 download

Generate cutouts:

snakemake -s Snakefile_cutouts --cores 1

Tip

Zenodo downloads are much faster (~minutes vs hours) and don’t require API credentials. Use this method unless you need years outside 2010-2024.

Directory Structure#

After installation, your directory should look like:

PyPSA-GB/
├── config/              # Configuration files
│   ├── config.yaml      # Active scenarios
│   ├── scenarios.yaml   # Scenario definitions
│   └── defaults.yaml    # Default values
├── data/                # Input data (versioned)
├── resources/           # Generated outputs
├── scripts/             # Python modules
├── rules/               # Snakemake rules
├── envs/                # Conda environment
├── Snakefile            # Main workflow
└── Snakefile_cutouts    # Weather data workflow

Troubleshooting#

Common Issues#

Conda environment creation fails

Try updating conda first:

conda update -n base conda
conda env create -f envs/pypsa-gb.yaml

Gurobi license not found

Ensure your license file is in the correct location:

Linux/macOS: ~/gurobi.lic
Windows: C:\Users\<username>\gurobi.lic

Snakemake not found

Ensure the environment is activated:

conda activate pypsa-gb
which snakemake  # Should show path in conda env

Next Steps#

Quickstart: Run Your First Model - Run your first model
Your First Custom Scenario - Create a custom scenario