Projects
Projects: Custom Search |
||
| Reference Number | NIA2_NGESO060 | |
| Title | FastOut | |
| Status | Completed | |
| Energy Categories | Other Power and Storage Technologies (Electricity transmission and distribution) 100%; | |
| Research Types | Applied Research and Development 100% | |
| Science and Technology Fields | PHYSICAL SCIENCES AND MATHEMATICS (Computer Science and Informatics) 30%; ENGINEERING AND TECHNOLOGY (Electrical and Electronic Engineering) 70%; |
|
| UKERC Cross Cutting Characterisation | Not Cross-cutting 100% | |
| Principal Investigator |
Project Contact No email address given National Grid plc |
|
| Award Type | Network Innovation Allowance | |
| Funding Source | Ofgem | |
| Start Date | 01 October 2023 | |
| End Date | 31 March 2024 | |
| Duration | ENA months | |
| Total Grant Value | £500,000 | |
| Industrial Sectors | Power | |
| Region | London | |
| Programme | Network Innovation Allowance | |
| Investigators | Principal Investigator | Project Contact , National Grid plc (100.000%) |
| Industrial Collaborator | Project Contact , National Grid plc (0.000%) |
|
| Web Site | https://smarter.energynetworks.org/projects/NIA2_NGESO060 |
|
| Objectives | The Project will be broadly delivered through 3 work packages (WPs):WP1: User research and design - Engage with ESO users to define the project problem statement and scope of the AI software solution, in particular, how the outcome of this project can provide outage advice to planning engineers.WP2: Data Science ingestion and analysis of relevant datasets (for example, outage requests, day-ahead constraints, wind forecasts) to determine an appropriate solution design to be tested with ESO. The project will continue to build a Proof-of-Concept (PoC) solution, which represents the core output of this phase.WP 3: Backend engineering actively identify and test potential deployment pipelines, such that models described above can be successfully integrated into ESO infrastructure within subsequently Beta/Live Phases.Using a combination of PowerFactory derived data for single circuit outages, historical outage versus constraint limit data and data on intact limits for each constraint boundary, Faculty will develop an AI tool that can assess a single outage at a given time against existing scheduled outages. Due to the nature of the data utilised in the project it will restrict how the data can be handled and shared; we have put mitigations in place with our security teams to ensure that the data used for the project follows a strict governance process with our supplier, and information that is deemed confidential will be redacted from the final reports and outputs to ensure we comply with our obligations. At the end of this project, we expect to have the following:Improved understanding of the relationship between outages and constraint limit loss.Proof of Concept tool for delivering a rapid assessment of the scale of cost of a planned outage.Agreed plan for how the tool will be deployed for regular use by end users, developed in close collaboration with ESO"s IT team.Established infrastructure for tool deployment as tested with a skeleton solution.The success of the project will be based on the following criteria:Benchmarking of the operability and estimated cost of an outage request provided by the tool relative to the operability and costs determined by engineers by running simulations in PowerFactory and calculating the costs of the outputs from the model outputs.Outage advice provided by a PoC solution, validated through acceptance testing with planning engineers in the NAP team and a wider ESO Steering Group. In line with the ENA"s ENIP document, the risk rating is scored Low.TRL Steps = 1 (3 TRL steps)Cost = 1 (£500k)Suppliers = 1 (1 supplier)Data Assumptions = 2Total = 5 (Low) Currently, most outage requests received by engineers in Network Access Planning (NAP) have to be assessed in PowerFactory to determine the likely impact on the network (PowerFactory is a custom power systems model of the electricity transmission network). The PowerFactory modelling is an interactive process and can take considerable time and requires significant expertise.The project would seek to better understand the relationship between outages and constraint limit losses. The relationship between outages, limit losses and other factors will be codified into a model that can calculate a cost indication (cheap vs. expensive) for scheduling an outage.This model will be the engine of a PoC tool to allow engineers to triage outage requests before decided whether to pursue more complex time-consuming analysis with PowerFactory. The project will also involve interviewing key stakeholders to determine, what "good enough" looks like and how the output from the modelling is best presented to users. These insights will be used to create a user-interface that is intuitive and fit-for-purpose.The overall solution has the potential to deliver substantial time-savings for planning engineers and other users as well as improve their understanding of the impact of outages on the network. This project aims to deliver a PoC AI solution to provide rapid outage request triaging. The PoC will be trained and tested on historical data but will not be deployable on ESO infrastructure. The specific objectives of the project are to develop:A PoC model that estimates the impact of an outage request on the network in terms of the operability of the request and the cost to the system if the request is approved and goes ahead. The estimate will be based on PowerFactory derived data for single circuit outages, historical constraint limit data and other features deemed to be relevant (e.g. wind forecasts).The project deliverables will include initial designs of a front-end user interface to represent how the bespoke software would be integrated into decision-makingBased on engineering discovery activities combined with user research, a plan for full deployment during a prospective Beta Phase (deployment out of scope of this phase) | |
| Abstract | Outage requests from asset owners are ideally scheduled when their effect on constraint limit loss is minimal. Currently, the network planning team model the effect of each outage on constraint limits using PowerFactory models. Although this process can account for the complexities of the network, it can be slow and contributes to a backlog of outage requests from asset owners, which can lead to costly emergency or short-notice outages from unresolved outage requests and results in penalties (Fail to Fly) from Ofgem.This project aims to explore the relationship between outages and constraint limit losses by examining historical outage and post-fault action data and developing an AI. model that can perform a rapid first pass assessment of outage requests reducing the search-space for time consuming PowerFactory modelling. | |
| Data | No related datasets |
|
| Projects | No related projects |
|
| Publications | No related publications |
|
| Added to Database | 02/10/24 | |
