Project Summary

The distribution connections queue is c.180GW and growing, driven by electrification of demand and growth in low-carbon generation and storage. Assessing overall network impacts from this ever-evolving connections stack is challenging, complicating networks' assessment of available headroom and future investment needs.

FastTrack, an Artificial Intelligence solution, simulates the impact of connection requests from primary substations to the Grid Supply Point. By analysing network capacity, existing load, connection data, and external drivers, it provides a consolidated view of future demand. This supports system planners in making informed decisions on investments, access products, and interventions to accelerate delivery and optimise capacity use.

Innovation Justification

Challenge:

FastTrack, aligned with Round 4, Challenge 1, is an AI-driven solution that simulates the impact of the distribution connection queue. It enhances system planning and headroom assessments, helping DNOs optimise existing network capacity and accelerate connection times.

Core Innovation:

New Capability: FastTrack offers a new capability to model the evolving connection queue's impact on the network. By capturing diverse connection behaviours, it gives system planners a dynamic, regularly updated view, moving beyond static forecasting methods.

Novel Application of AI Techniques: Using a novel combination of probabilistic forecasting and simulation, FastTrack quantifies uncertainty in connection uptake and behaviours. This moves beyond traditional methods like DFES or individual assessments, offering system-wide insight into likely network impacts.

Surpassing the State-of-the-Art:

Forecasting at System Level: FastTrack models the impact of individual/grouped connection requests from substations to GSP. This offers visibility of how local decisions shape wider system outcomes; capability not currently available in existing tools.

Dynamic Risk Assessment: FastTrack provides a probability-weighted view of the evolving queue up to the GSP, helping DNO system planners to make more robust decisions on accepting connection requests and designing optimal interventions (e.g., reinforcement, flexibility, and access products).

Faster, More Flexible Planning: FastTrack's capabilities will enable DNOs to more quickly design and crucially adapt interventions (e.g., prioritisation of investment and design of access products). Unlike current methods that can take days or weeks to develop headroom assessments and often need to be recalculated as the queue changes, FastTrack provides an up-to-date assessment of operating headroom, allowing DNOs to quickly adapt plans as the queue changes.

Working in the Open: During the Discovery phase, the project engaged openly with stakeholders across the DNO business. Their input and feedback directly shaped the direction of the Alpha project.

Building on Previous Research:

The Innovation Appendix maps the proceeding innovations focused on accelerating the connections queue and how FastTrack fits in.

Past projects have improved parts of the connection process, e.g. UKPN's HV Auto Quote for quote generation; ENA ConnectDirect and NGED Click2Connect for application efficiencies; NPg Artificial Forecasting for load forecasting. However, none assess cumulative, risk-weighted impacts across the GSP. FastTrack builds on these efforts to deliver a fundamentally new planning tool.

TRL, IRL, CRL:

TRL: Discovery reached TRL2, validating the solution concept with system planners. At Alpha, a streamlined PoC build focusing on four solution elements will raise this to TRL4.

IRL: Discovery formulated the business concept (IRL2), with Alpha seeking to validate the value proposition and develop a firm route to market (IRL4).

CRL: Discovery has considered the commercial proposition (CRL2), to be developed to CRL4/5 (early market engagement) throughout Alpha, working with DNO partners to facilitate longer-term adoption across the sector.

Size/Scale: The scale of the project, being inherently large/risky, is proportional to the requirements to rigorously develop and test AI solutions and determine a clear route to adoption.

Funding/BAU: SIF funding is crucial as the ambitious scope, low TRL and inherent risks of this solution cannot be advanced under BAU. The benefits from introducing AI methods will be magnified by the foundational work undertaken within this SIF project, which develops capability at speed and rigour over and above current plans, potentially unlocking further related innovations, complementing DNO's Digitalisation Strategies and Action Plans.

Counterfactuals: Expanding current resources or repurposing existing tools will not be sufficient to meet the growing demand in connection requests and interventions (e.g., bottom-up substation-by-substation analyses). These tools lack the ability to assess cumulative, forward-looking impacts and are limited to substation-level analysis. In contrast, FastTrack's AI-driven approach is scalable across SSEN and other DNOs incorporating diverse data inputs more readily than current methods.

FastTrack Alpha- Innovation Justification_Final.pdf (opens in a new window)

(/application/10166254/form/question/47959/forminput/135591/file/833307/download) .

Impacts and Benefits

Pre-Innovation Baseline and metrics: DNOs currently lack the capability to assess, in real-time, the impact of the evolving connections queue on the electricity network, from primary substations up to the Grid Supply Point. Planning decisions rely heavily on manual, static methods that cannot keep pace with the scale and volatility of new connection requests, driven by electrification. Key metrics defining the baseline include:

*Planned connections-related reinforcement expenditure at HV and above (from SSEN's ED2 business plans, projected forward to 2038).

*Time to Quote (TTQ) and Time to Connect (TTC), segmented by voltage level and asset type.

*Growth in system planning FTE that would otherwise be required, under existing approaches, to keep pace with the scale of interventions and connections requests

Annual improvements (c.2% pa) in TTQ/TTC are expected under BAU due to wider connections reform and other initiatives ongoing at SSEN.

Forecast Net Benefits to Consumers:

If implemented into business-as-usual at SSEN (Option 1), FastTrack would deliver an NPV of c.£35 million by 2038, assuming operational deployment from2028. It achieves this through three core benefit channels:

Financial: Reduced Cost of Network Operation

Channel 1. More efficient planning and delivery of interventions. FastTrack provides system planners with an ongoing view of the load impact of the connections queue, facilitating better prioritisation of network interventions. This allows interventions to be scoped faster and more flexibly as the queue evolves.

This is quantified through a 2% efficiency saving on 'relevant' connections related reinforcement, defined as:

*In-scope: 55% of reinforcement applies to HV and above, the portion of the network targeted by FastTrack.

*Relevant portion: 70% of 'in-scope' investment is likely targetable by FastTrack; this is the portion without transmission dependencies (39% of projects are currently transmission-dependent (ENA), however following updates to CMP446, raising the threshold that transmission impact assessments must be conducted, we assume 30%). Channel 2. Operational Efficiency: avoided Headcount Growth

Under BAU, SSEN would require c.10 additional FTEs by 2035 to support manual queue analysis and planning. FastTrack automates core forecasting tasks, enabling FTE cost avoidance while improving performance. This benefit is estimated in the range of £6--8 million in avoided costs over the forecast period (pre-discounting).

Revenues: improved access to revenues for users of network services

Channel 3. Expedited revenues to developers. Enabling SSEN system planners to better and more flexibly design and execute interventions streamlines Time to Quote (given this is where the majority of system planning effort lies) and Time to Connect (given higher quality and more streamlined interventions). Given FastTrack will first focus on demand, the CBA considers the impact of these efficiencies on revenues for demand-sites (data centres and commercial sites) and distribution-connected storage only.

Expedited revenues are calculated using typical assets and revenue profiles. E.g. for batteries, a 10MW battery size is assumed, with monthly revenues collected from market analysts Modo energy. For each asset:

*Days saved in TTQ/TTC are estimated based on the potential impact of FastTrack -- these typically reflect a c.5% time reduction relative to the baseline. Time to Connect impact assumptions are half those of Time to Quote.

*Days saved are multiplied by typical revenue per day, to aggregate expedited revenue access by asset type, multiplied by the expected number of connections from DFES.

Benefits Realised to Date

FastTrack has already delivered early-stage value by:

*Demonstrating potential for an AI solution to enhance system planning

*Showcasing the value in aggregating disparate public/private data sources into a single solution.

*Engaging system planners to validate early prototypes and shape user-centric design. These steps have reduced technical risk and created strong foundations for build of the proof-of-concept at Alpha.