Data is one of the biggest enablers of a decarbonised, decentralised and digitalised future energy system. It is an essential tool to keep the cost of transforming our energy system as low as possible.
While MEDA aimed at creating a common data architecture, the Modernising Energy Data Applications (MEDApps) competition funded the development of data applications that address the needs of users and developers of local energy systems. These projects are required to provide scalable commercial opportunities, and to integrate with the Open Energy data architecture, the winning project of the MEDA competition.
The MEDApps competition is structured as a two-phase competition.
Phase 1 covers the discovery and alpha phase, running from April 2021 until July 2021.
Phase 2 covers the beta phase, running from September 2021 until June 2022.
Provide tailored support to the projects for testing their solution, liaising with industry stakeholders, and identifying links with other initiatives,
Identify and help projects resolve wider system issues such as access to network operators data,
Facilitate knowledge sharing, by coordinating events and stakeholders engagement.
MEDApps phase 1
In the first phase of the competition, the programme funded nine projects covering the cost for their discovery and alpha phase. A range of new applications was proposed, with each of the projects dealing with one of the biggest challenges of the energy transition such as electric vehicles’ uptake, facilitated design of district heating networks, local area energy planning, indoor farming, gas meter leaks detections, and fuel poverty.
Energy Systems Catapult assisted the nine projects in their user research by introducing the projects to potential users and initiatives in the sector, reviewing their business model, and identifying data sources and supporting the negotiations with data providers.
Supporting fuel poor households through data integration and AI – led by UrbanTide
Being a two-stage competition, only five out of nine projects were admitted to the second round, funding the development of the beta phase of these data applications. In the nine months of this competition, the Energy Systems Catapult will provide a wide range of tailored support to these projects from usability testing to overcoming challenges related to data acquisition. Read more about the five phase-2-winning projects in the sections below.
Aiming to assess the value of a Local Area Energy Planning (LAEP) using big data analysis to help non-technical energy planners to rapidly model energy scenarios in close to real-time using a novel common analysis framework.
The application should build on the common data architecture from recent data projects – accessing, integrating and exploring the datasets from those projects. It should also integrate open datasets that are currently too dissipated and difficult for planners to explore without expensive specialist support.
The service vision is a web GIS single page application which enables users to view, download/upload, and analyse building-level data in order to plan and optimise the siting of any combination of low carbon technologies at scale.
By maintaining this data in up-to-date and geospatially mapped formats, accessible on web-based visualisation platforms, we remove data-as-a-barrier to LAEP analysis. Better whole-system planning with full stakeholder engagement has been shown to reduce cost of capital by 7% saving the UK taxpayer £1.4 billion annually in infrastructure expenditure.
Aiming to build a generative design tool that will automate the optimal design of district heating networks. Currently, design is costly and time-consuming and often lacks sufficiently detailed data to ensure system efficiency. In many cases this leads to over-sizing of equipment which increases capital costs which are ultimately passed on to the end consumer or can make projects unviable.
An automated, computer-driven process, will radically reduce the up-front costs of modelling and system sizing;
Using the latest AI techniques combined with detailed geographic data will optimise system design to increase the affordability of heat networks and maximise their viability.
End consumers will benefit through lower heating costs, warmer homes/buildings and reduced fuel poverty. Taking into account multiple design options will allow for greater focus on zero-carbon fuel sources (including hydrogen) which will promote a new generation of true zero-carbon district heat.
This would help the UK to meet its 2050 vision of meeting 20% of heating demand by heat networks.
Alian (Farad.ai) – AI for low carbon technology site optimisation
Aiming to model the shift of complex power flows in the electricity network, to act as a central nervous system for the power sector. The platform uses cutting-edge analytics to understand and predict strains on the grid to make use of congested electricity and incorporate more renewable energy.
The strains on the electricity network are going to be more severe in time as the world starts to deliver their Net-Zero ambition, which means the same power network will have to deliver 3x as much electricity to consumers.
Farad.ai envisions a future where the energy system will be working in harmony with other utilities, including gas and water networks, to revolutionise the energy sector.
Mind Foundry – Energy-focused geospatial system using multi-sectoral data to deliver net zero
Aiming to use geospatial data, which combines location information with attribute and often temporal information, to contribute to the evolution of the distributed energy paradigm.
The ability to analyse sparse or incomplete data, including from smart meters, electric vehicle (EV) uptake, EV charging type, locations and user profile, has a range of implications that are difficult to model with conventional approaches.
Mind Foundry are building an energy-focused geospatial system that will enable the user to visualise overlays of multivariate spatially and temporally varying data, model and predict trends and correlations, infer across areas of sparse data collection, and model the effects of changes on the system such as varying supply, demand or infrastructure. It will further allow for the simulation and testing of different strategies, for example, alternative charging point placement.
Using world-class expertise in Bayesian optimisation and scalable probabilistic models to significantly improve the accuracy of AI models. The successful application of this approach will enable the local energy sector to be more quantitative and targeted in its planning and prioritising of resources.
Urbantide – Supporting fuel poor households through data integration and AI
Aiming to combine for the first time, UK-wide smart meter system metadata with multiple cross-sector data sources to identify households in fuel poverty in a completely novel way and those that would benefit from energy efficiency programmes. Propose interventions that can maximise energy reduction toward Net Zero whilst reducing fuel poverty, and support better use of data and advanced statistics /machine learning in delivering benefits to the fuel poor.
This builds on outstanding results from Phase 1 which included:
The first-ever sharing of 18 months of synthetic/ anonymised DCC smart meter metadata from 11 million meters,
Analysis uncovering patterns of behaviour that reveal novel insight on fuel poverty, proving the feasibility and theoretically illustrating better outputs than the status quo approaches for fuel poverty identification,
Confirmation of demand for the uSmart:ZERO service through engagement with key stakeholders and prospective customers,
New ethics and Data Governance approvals obtained for SERL data to be accessed by Accredited Researchers (UCL) for fuel poverty research.