Data Study Group - January 2025

Introduction

Stage 1: Precursor Stage (part-time, online)

• The Precursor Stage runs Monday 20 January – Friday 24 January 2025, in the lead up to the Event Stage.
• The maximum time commitment is 2.5 hours a day.
• This includes online workshops, presentations and team building which will prepare participants for the Event Stage.

Stage 2: Event Stage (full-time, online)

• The Event Stage runs virtually over two weeks, Monday 27 January – Friday 7 February 2025.
• The core working hours will be 09:00 - 17:00 GMT every week day. Flexibility will be demonstrated regarding those participating in different time zones, however you will need to be available for part of the GMT day to allow some crossover work time with your GMT based team members.
• Group work begins and continues throughout.

Applicants should be able to commit to the duration of the event. The Alan Turing Institute is committed to supporting supporting individual circumstances, please do not hesitate to email [email protected] to discuss any reasonable adjustments.

Challenges

Artificial Intelligence for Decarbonisation's Virtual Centre of Excellence (ADViCE)

Understanding Heat Pump Performance: Improving Efficiency and Adoption.

 

Heat pump efficiency in real-world use varies dramatically, with performance ranging from 150% to 450%. The drivers of underperformance in each case can be understood by experts, but this is difficult to scale.

This challenge seeks to use monitoring data to automatically identify issues affecting individual heat pumps, particularly those that can be addressed through system configuration, installation quality, and control strategies.

As part of the DESNZ Electrification of Heat trial, Energy Systems Catapult collected monitoring data from over 740 homes, including 2-minute resolution heat pump data, property characteristics, installation details, and demographic information.

This study aims to use this data to identify ways to quickly diagnose underperformance and recommend targeted interventions, contributing to the wider adoption of heat pumps and the UK's Net Zero goals.

British Geological Survey (BGS)

Detecting shallow gas from seabed seismic images.

Shallow gas below the seabed is a hazard for crucial NetZero infrastructure such as wind turbines. Quantifying the structure of the subsurface is important to understand the load-bearing capability of locations and gas can mask soil units, leading to uncertainty in ground conditions. There are limited maps available for shallow gas, despite their importance for seabed planning and development. New surveys to capture the information required to create maps of shallow gas are extremely expensive, so it is important to capture as much information as possible from existing information. 

BGS holds many sources of legacy information including hundreds of thousands of kilometres of seismic images, dating from 1966-2012. These data were mostly recorded directly on paper and have been stored as physical records until digitisation into scanned records. Manual checking of the records for evidence of shallow gas (such as gas chimneys, acoustic blanking or bright-spots) is unfeasible for such large numbers of images, so an automatic data-driven method is required. Machine learning techniques will allow for shallow gas to be identified in legacy records, and geographically located. This crucial information about shallow gas hazard will support successful siting of offshore renewable infrastructure.
 

The Centre for Postdoctoral Development in Infrastructure, Cities and Energy (C-DICE)

Re-purposing and decommissioning energy assets to maximise the impact on net zero carbon emissions.

The UK government’s commitment to achieving net zero greenhouse gas emissions by 2050 will require a highly skilled workforce, with expertise in advanced technologies, sustainable practices, and innovative solutions across key sectors such as energy, transport, and industry to significantly reduce the nation’s carbon footprint.

In collaboration with industry partners, C-DICE has identified critical areas for development to maximise the contributions and impact of doctoral and postdoctoral-level skills in achieving net zero by 2050. However, before we can pinpoint the future high-level skills needed, we must first better understand what that future will entail.

The DSG challenge will draw on data from two decarbonisation scenarios for the UK energy system, developed by the Energy Systems Catapult (ESC), along with open-access data on power plants. This data will be used to gain insights into how decommissioning and repurposing energy-related assets can best support progress towards the net-zero target. 

The challenge aims to build an optimisation framework for evaluating solutions to decommission and repurpose energy-related assets, with an emphasis on maximising the UK’s progress toward its 2050 net-zero goals. Additionally, it will generate a set of potential solutions that could be used to help identify critical skills gaps in the decommissioning process. While participants are not expected to provide direct information on skills gaps, their recommendations in this area will be welcomed, if relevant.

ScotRail

Understanding rail and road user behaviour to reduce car miles in Scotland.

The Scottish Government has set targets to reduce the number of car miles driven in Scotland by 20% by 2030 to support net-zero ambitions.  The vast majority of car miles driven nationally arises from longer trips, many of which have the potential to be replaced by rail journeys.

In this DSG, data related to both rail and road usage will be analysed to uncover new behavioural patterns and to provide evidence to support investment to incentivise a switch away from car journeys to rail.  Sources of data will include passenger counts, timetabling data, network topologies, rolling stock capacities, disruption data for the rail network along with detailed trunk road traffic movements including location, vehicle type, speed and direction.

Better understanding of both rail and road user behaviour, such as the effect of disruptions (roadworks, cancellations, adverse weather) will support strategic decisions in improvements in timetabling, travel incentivisation schemes, improved rolling stock maintenance leading to an increase in rail travel adoption, and an improved rail service.  Increased rail travel as an alternative to car travel will lead to a reduction in car miles driven, supporting governmental net zero ambitions.