Tutorials

Interactive Jupyter notebook tutorials demonstrating PyPSA-GB’s network-constrained optimal power flow (LOPF) capabilities across historical and future scenarios.

Scenario Tutorials

• Historical Baseload
• Historical Renewables Era
• Future Scenario: Holistic Transition (2035)
• Future Scenario: Electric Engagement (2050)

Component Tutorials

• Networks
• Demand
• Generators
• Marginal Costs
• Renewable Generation
• Storage Systems
• Interconnectors
• Hydrogen Systems
• Heat Pump Flexibility
• Electric Vehicle Flexibility
• Demand Side Response
• Market Dispatch

Tutorial Overview

These tutorials demonstrate PyPSA-GB’s core capability: network-constrained optimal power flow (LOPF) modeling across different time periods and network resolutions.

Scenario Tutorials (1-4)

Complete scenario walkthroughs from network building to results analysis:

#	Notebook	Year	Network	Solve Time	Focus
1	Historical Baseload (2015)	2015	Reduced (32 buses)	5-10 min	Traditional dispatch, coal/nuclear dominance
2	Historical Renewables (2023)	2023	Full ETYS (~2000 buses)	30-60 min	High renewable penetration, network detail
3	Future Near-Term (2035)	2035	ETYS clustered (100 buses)	10-15 min	Renewable dominance, storage, network upgrades
4	Future Long-Term (2050)	2050	ETYS clustered (100 buses)	10-15 min	Near-100% renewables, stress test, adequacy

Component Tutorials (5-16)

Deep dives into specific aspects of power system modeling:

#	Notebook	Topic	Description
5	Networks	Topology & Power Flow	Buses, lines, transformers, coordinate systems
6	Demand	Load Modeling	ESPENI/eload profiles, temporal patterns, spatial distribution
7	Generators	Dispatch Optimization	Thermal/renewable generators, merit order, capacity factors
8	Marginal Prices	Price Formation	LMPs, congestion pricing, generator revenues
9	Renewables	Variable Generation	Wind/solar profiles, curtailment, capacity factors
10	Storage	Flexibility Resources	Batteries, pumped hydro, state of charge, arbitrage
11	Interconnectors	Cross-border Trade	HVDC links, import/export flows, utilization
12	Hydrogen	Sector Coupling	Electrolysis, H2 storage, hydrogen-to-power
13	Heat Flexibility	Heat Pump Demand	TANK/COSY modes, COP, thermal storage, demand shifting
14	EV Flexibility	EV Smart Charging	GO/INT/V2G tariffs, battery storage, Vehicle-to-Grid
15	Demand-Side Response	Event Response	Saving Sessions, demand reduction events, DSR capacity
16	Market Dispatch	Wholesale + BM	Copperplate wholesale dispatch, balancing redispatch, ELEXON/NESO validation

Running Tutorials

Prerequisites

1. Complete the Installation

2. Activate the environment:

   conda activate pypsa-gb
   

3. Generate the required scenario networks (see below)

Generate Scenario Networks

Before running the tutorials, build the required networks:

# Notebook 1: Historical 2015 (Reduced network)
snakemake resources/network/Historical_2015_reduced_solved.nc -j 4

# Notebook 2: Historical 2023 (Full ETYS - takes longer)
snakemake resources/network/Historical_2023_etys_solved.nc -j 4

# Notebook 3: Future 2035 (Clustered ETYS)
snakemake resources/network/HT35_clustered_solved.nc -j 4

# Notebook 4: Future 2050 (Clustered ETYS)
snakemake resources/network/EE50_clustered_solved.nc -j 4

The market dispatch tutorial reads solved market and validation outputs rather than a single solved network. Build the relevant market validation targets first, for example:

snakemake resources/analysis/Validation_Jan2020_UniformNetworkBaseline_bm_validation.html -j 4
snakemake resources/analysis/Validation_Jan2020_UniformNetworkBaseline_neso_validation.html -j 4

Launch Jupyter

# Navigate to tutorials directory
cd docs/source/tutorials

# Start Jupyter
jupyter notebook

First-Time Setup

Install the kernel if needed:

python -m ipykernel install --user --name=pypsa-gb

Then select the pypsa-gb kernel in Jupyter.

Tutorial Requirements

System Requirements

Notebook	RAM	Solve Time	Disk Space
1 - Historical 2015	4 GB	5-10 min	~500 MB
2 - Historical 2023	16 GB	30-60 min	~2 GB
3 - Future 2035	8 GB	10-15 min	~1 GB
4 - Future 2050	8 GB	10-15 min	~1 GB
16 - Market Dispatch	16 GB+	Depends on market scenarios	Several GB for monthly validation outputs

Solver Requirements

Tutorials require Gurobi solver (free academic license available) or HiGHS (open-source).

See Installation for solver setup.

Contributing Tutorials

We welcome new tutorial contributions:

1. Create a notebook in docs/source/tutorials/

2. Add to the toctree in this index (in appropriate section)

3. Test that it builds correctly

See Contributing for guidelines.