FastFlow
Project summary
FastFlow investigates whether physics-constrained artificial intelligence can accelerate Alternating Current power flow and optimal power flow studies used for transmission planning and operational decision support.
| Name | Status | Project reference number | Start date | Proposed End date | Expenditure |
|---|---|---|---|---|---|
| FastFlow | Live | NIA3_NESO012 | Jan 2027 | Jan 2029 | £480,000 |
FastFlow investigates whether physics-constrained artificial intelligence can accelerate Alternating Current power flow and optimal power flow studies used for transmission planning and operational decision support. The project will develop graph neural network and Transformer-based surrogate models using IEEE benchmark systems and GB transmission data. Models will embed power-system physics and operational constraints and be tested under varying network conditions. Work is organised into four packages and outputs include validated datasets, trained models, evaluation reports, user interface for model interaction, user guidance and a final report. Performance will be benchmarked against conventional PF and AC-OPF solvers for accuracy, runtime, scalability and topology generalisation. Risks include data availability, quality and model performance. These will be managed through validation, testing and comparison with established tools. Outputs are intended to complement existing solvers.
Benefits
1. What is the expected benefit delivered directly by THIS PROJECT?
1a.Short outcome statement (clear and specific)
FastFlow will develop and validate AI-based power flow (PF) and optimal power flow (OPF) models that enable much faster network studies, helping NESO assess more scenarios and support faster, better-informed planning and operational decisions.
Benefit type: Knowledge - Operational efficiency, improved decision-making, and enhanced network planning capability.
1b. Who will benefit from this project and in what way?
NESO operational and planning teams (including Offline Modelling, Network Operability and Network Planning teams) will benefit from validated knowledge, datasets, benchmarking results and prototype analytical tools that can potentially reduce the time required to perform PF and OPF studies
The wider GB electricity sector will benefit from openly disseminated learning regarding the application of AI techniques to large-scale transmission network analysis, helping establish whether these methods are suitable for future operational and planning use cases.
1c. How confident are you this project will deliver this benefit?
Medium
The project's primary output is knowledge generation through development, validation and benchmarking. Even if the final AI models do not achieve deployment-ready performance, the project will still provide valuable evidence regarding the suitability, limitations and future requirements of AI-assisted PF and OPF approaches for GB transmission system.
1d. Can this benefit be quantified? (For development and demonstration projects provide quantified estimates where reasonably possible. If quantification is not currently feasible, explain why and what evidence supports the expected benefit)
Yes - provide metric, baseline, expected improvement
No - explain why not at this stage
Yes. The direct project benefit will be quantified through measurable technical performance metrics including:
FastFlow will be assessed against conventional PF and OPF solutions, targeting voltage magnitude error ≤0.005 p.u., voltage angle error ≤1°, OPF optimality gap <0.5%, ≥99% constraint compliance, successful operation on unseen GB network topologies, and study runtimes approximately 2-5 times faster than conventional approaches.
These metrics will be validated against conventional PF and OPF solutions throughout the project.
Future / Full-Scale Benefit (if applicable - think beyond direct project outputs, e.g. operational rollout, industry adoption, policy influence, future cost reduction, resilience improvement)
Apply relevant category (Knowledge, Social, Environmental, Financial)
2. What is the expected benefit if this idea is implemented at full scale?
2a. High-level outcome statement
If successfully deployed within NESO workflows, FastFlow could substantially reduce PF and OPF study runtimes, enabling significantly more operating scenarios, contingencies, outages and network configurations to be assessed within existing planning and operational timescales while maintaining acceptable engineering accuracy
2b. Who will benefit from this project in the future and in what way?
Link benefit - stakeholder clearly
NESO engineers and planners
· Reduced time spent on repetitive PF and OPF calculations (These will be validated against conventional PF and OPF solutions throughout the project).
· Increased study throughput.
· Faster assessment of connection requests, outages, switching actions and contingency studies.
Electricity consumers
· Financial benefit: Potentially lower operational and planning costs through more efficient use of engineering and computational resources.
· Knowledge benefit: Improved operational decision support and system resilience
2c. What needs to happen for this future benefit to be realised?
e.g. rollout, policy change, integration, further development
The following would be required:
· Successful achievement of project performance targets.
· Validation against NESO operational requirements.
· Integration with NESO data sources and modelling workflows.
· Completion of cyber-security, model assurance and governance processes.
· User adoption, training and operational deployment.
· Continued benchmarking against conventional PF and OPF tools during initial rollout.
2d. How confident are you this project will deliver this benefit?
Low to medium
The project is specifically designed to address significant technical uncertainties relating to model accuracy, robustness, scalability and topology generalisation. While the project is expected to generate valuable learning, successful large-scale operational deployment depends on demonstrating acceptable performance on GB-scale operational datasets
2e. Can this benefit be quantified? (For development and demonstration projects provide quantified estimates where reasonably possible. If quantification is not currently feasible, explain why and what evidence supports the expected benefit)
Yes - provide metric, baseline, expected improvement
No - explain why not at this stage
Technical benefits can be quantified through:
· Approximately 10× reduction in PF and OPF study runtime.
· Ability to assess approximately 10× more scenarios within the same computational window.
· PF accuracy targets of 0.005 p.u. voltage magnitude MAE and 1° voltage angle MAE.
· OPF average optimality gap below 0.5%.
· ≥99% operational constraint satisfaction
Future financial benefits cannot yet be robustly quantified because they depend on:
· Actual NESO study volumes.
· Future operational deployment costs.
· Integration and maintenance costs.
· Measured runtime improvements achieved during the project.
| Name | Published |
|---|---|
| NIA Project Registration and PEA Document | 03 Nov 2026 |