Comprehensive analysis of electricity flexibility market activations across different countries/regions.
Author: Emil Mahler Larsen, CTO, Utiligize Date: October 2025
This project analyzes flexibility dispatch data from Distribution Network Operators (DNOs) and Transmission System Operators (TSOs) to understand:
- When and where flexibility is dispatched
- What technologies provide flexibility services
- Market dynamics, pricing, and provider competition
- Tender success rates and procurement efficiency
- Comparison between different market models
Status: β Complete with full analysis (23,607 dispatches)
UK Power Networks (UKPN) flexibility market - complete DNO-level analysis.
Data Source: UK Power Networks Open Data Portal
Key Findings:
- Β£13.1M market value, 28.5 GWh dispatched (Apr 2023 - Oct 2025)
- 10% tender success rate (90% of zones unactivated)
- 56% EV charging (predominantly home chargers aggregated by retailers)
- 48% monopoly zones (single provider only)
- 158 active zones, 31 providers, 439 flexible units
- Distribution-level: 86% at 33/11kV substations
- Top provider: Octopus Energy (Β£3.2M), followed by Γrsted (Β£2.3M wind)
- Contract length: 6.4 months average (seasonal/annual mix)
- Jackpot events: 15% of revenue from Β£4k-40k/MWh peak prices
π See GB folder for complete analysis
Status: β Complete analysis (959 clearing events)
GOPACS (Grid Operators Platform for Congestion Solutions) flexibility market data.
Data Source: GOPACS Public API
Key Findings:
- 959 clearing events, 14,912 PTU records (Dec 2018 - Oct 2025)
- TSO-dominated: TenneT 99.9% of volume (transmission-level)
- DSO minimal: Only 0.1% of market (682 MWh over 7 years)
- 4 operators: TenneT (TSO), Liander, Enexis, Stedin (DSOs)
- 15-minute PTUs (standard Dutch market structure)
- Peak 2021-2022, significant decline since 2023
- Average volume: 145.6 MW per PTU (transmission-scale)
Comparison: NL is transmission-focused (TSO), GB is distribution-focused (DNO)
- NL DSOs: 100 MWh/year
- GB UKPN alone: 11,193 MWh/year (25% of GB market)
- UKPN is 112x larger; Full GB estimated ~450x larger than NL DSOs
π See NL folder for complete analysis
| Aspect | π¬π§ Great Britain (UKPN) | π³π± Netherlands (GOPACS) |
|---|---|---|
| Market Focus | DNO distribution-level | TSO transmission-level |
| Avg Capacity | 0.44 MW | 145.6 MW |
| Dominant Tech | EV home chargers (56%) | N/A (cleared buckets) |
| Market Type | Retailer-aggregated residential | Grid-level congestion |
| Granularity | Hourly dispatches | 15-minute PTUs |
| DSO Activity | Very active (28.5 GWh) | Minimal (0.7 GWh) |
| Trend | Growing 4x (2023-2025) | Declining (peak 2021-2022) |
Key Insight: GB leads Europe in distribution-level flexibility markets, while NL focuses on transmission-level congestion management.
uk-open-data/
βββ .gitignore # Excludes CSV data files
βββ README.md # This file
βββ GB/ # Great Britain (UK) analysis
β βββ README.md # GB-specific documentation
β βββ ANALYSIS_SUMMARY.md # Complete findings report
β βββ DETAILED_FINDINGS.md # Q&A on specific topics
β βββ CONTRACT_STRUCTURE_EXPLAINED.md
β βββ TENDER_STRUCTURE_EXPLAINED.md
β βββ POST_CORRECTIONS.md
β βββ fetch_flexibility_data.py
β βββ fetch_all_data.py # Bypass 10k API limit
β βββ analyze_flexibility_market.py
β βββ analyze_tender_success.py
β βββ analyze_peak_pricing.py
β βββ visualize_flexibility_market.py
β βββ create_square_graphs.py
β βββ requirements.txt
β βββ *.png # 10+ visualization outputs
β βββ square_graphs/ # 10 SVG graphs (1000x1000px)
βββ NL/ # Netherlands analysis
βββ README.md # NL findings and comparison
βββ fetch_gopacs_data.py
βββ analyze_gopacs_market.py
βββ visualize_gopacs_market.py
βββ requirements.txt
βββ *.png # 5 visualization outputs
Both country analyses provide:
- Temporal Patterns: Monthly, weekly, daily, and hourly dispatch trends
- Geographic Distribution: Zones with highest flexibility needs
- Technology Breakdown: Asset types providing flexibility
- Market Dynamics: Provider competition and market concentration
- Pricing Analysis: Cost trends by technology, product, and dispatch type
- Success Metrics: Market effectiveness and utilization rates
- Growth Trajectory: Market development over time
pip install pandas numpy matplotlib seaborn requestsGB (UKPN): Register at UKPN Open Data Portal, then set:
export UKPN_API_KEY=your_key_hereNL (GOPACS): No authentication required - public API
cd GB/
# Set API key
export UKPN_API_KEY=your_key_here
# Fetch all data (bypasses 10k limit with date filtering)
python fetch_all_data.py
# Run analysis
python analyze_flexibility_market.py
python visualize_flexibility_market.py
python analyze_tender_success.py
python analyze_peak_pricing.py
# Or generate publication graphs (10 square SVGs, 1000x1000px)
python create_square_graphs.pycd NL/
# Fetch data (no API key needed)
python fetch_gopacs_data.py
# Run analysis
python analyze_gopacs_market.py
python visualize_gopacs_market.py- DNOs/TSOs: Benchmark flexibility procurement efficiency across countries
- Flexibility Providers: Market intelligence and opportunity identification
- Researchers: Study DSO vs TSO flexibility market evolution
- Policymakers: Compare market designs (GB's DNO model vs NL's TSO model)
- Investors: Evaluate flexibility service opportunities
- Energy Retailers: Understand residential EV aggregation potential
- Distribution-focused (DNO level, 33/11kV substations)
- Home EV aggregation via energy retailers (Octopus, OVO, Ohme)
- 56% residential charging paused during peak demand
- Growing rapidly (4x growth 2023-2025)
- High failure rate (90% of tenders unsuccessful)
- Jackpot pricing (Β£20-40k/MWh during winter emergencies = 15% revenue)
- Transmission-focused (TSO level, TenneT)
- Grid-level congestion management
- DSO activity minimal (0.1% of market)
- Declining trend (peak 2021-2022, down 93% since)
- 15-minute PTUs (standard Dutch market structure)
GB's distribution-level market is vastly more developed than NL's:
- UKPN alone (25% of GB): 112x larger than all NL DSOs combined
- Full GB estimated: ~450x larger than NL DSOs
- NL DSOs activate only 100 MWh/year vs GB's ~45,000 MWh/year
This shows GB's clear leadership in DNO-level flexibility services, while NL focuses on TSO transmission-level congestion.
- GB: CC BY 4.0 (UK Power Networks)
- NL: Public data (GOPACS platform)
To add a new country/region:
- Create a new folder (e.g.,
DE/for Germany,FR/for France) - Add README with data source and API details
- Implement fetch, analyze, and visualize scripts
- Document findings in ANALYSIS_SUMMARY.md
- Update this main README with key findings
Analysis and visualizations ready for:
- Social media posts (10 square SVG graphs in GB/square_graphs/)
- Academic papers
- Policy reports
- Market intelligence briefings
- Presentations and conferences
Contact Emil Mahler Larsen (Utiligize) for collaboration or data sharing.
Project Started: 2025-10-21 Last Updated: 2025-10-21 Datasets: GB 23,607 dispatches | NL 959 events Coverage: GB Apr 2023 - Oct 2025 | NL Dec 2018 - Oct 2025