In November 2020, the Centre for AI & Climate (CAIC) organised a workshop on AI for Net Zero Electricity. The workshop was attended by over 35 participants from the electricity sector, including representatives from Ofgem, Energy Systems Catapult, National Grid, SSE, ScottishPower, EDF, BP, UK Power Networks, UCL, E.On Future Energy Ventures, and others.
The CAIC team are grateful to all of the participants who contributed their time and ideas before, during and after the workshop.
The workshop focussed on the opportunities and challenges faced by participants organised around the following topics:
Future vision of AI in the electricity system
Data Governance
Grid balancing
Moving from concept to deployment
Building data science capabilities
Overall, all parties were optimistic about the potential benefits of digitalisation and recognised its potential role in reaching a net zero electricity system. Read on for a summary of the workshop discussions and its resulting recommendations for actions to address some of the issues identified.
High level outcomes
There was a high level of alignment across the sector on the need to decarbonise and digitalise the electricity system and awareness of the potential benefits of data science in this process. While different businesses are at different stages in their digitalisation journey, many have made progress in developing their foundational data governance procedures and basic data science capabilities.
There is uncertainty regarding the role of AI in the future electricity system. Although there are many potential use-cases, AI applications remain at an early stage of development and deployment, and there is no shared view of what the grid will look like in 5-10 years’ time. It is therefore difficult for businesses to plan and receive funding for their future digitalisation activities.
Recommendation 1: Develop a vision and roadmap for the role of AI in reaching a net zero electricity system, in consultation with businesses and subject matter experts. This should feed into the government’s proposed energy data and digitalisation strategy.
Internal cultural barriers to digitalisation was a topic that came up many times. Energy businesses have been traditionally engineering-focused, with a reliance on staff who make decisions based on deep technical expertise. There is therefore a cultural barrier to digitalisation and data science, with mistrust in the outputs of “black-box” AI models, a reluctance to overhaul legacy IT and data governance systems, and a lack of understanding of the value of collecting and using data.
Recommendation 2: Create and distribute energy-AI training resources to improve data literacy across the sector, and build links with universities to attract data science talent.
Recommendation 3: More research is needed into explainable AI to improve trust and usability of AI applications in the energy sector.
Although there has been progress in improving sector-wide data governance through initiatives such as the ENA data working group and the Modernising Data Access competition, there is an appetite for improved guidance on sector-wide data standards (e.g., meta-data standards and SIMs). Open data is widely identified as a key enabler for sectoral collaboration in developing improved and larger-scale AI applications.
Recommendation 4: Coordinate the development of data standards and prioritise access to key datasets, building on existing initiatives in the sector.
Finally, although it is recognised that Ofgem and BEIS are well placed to lead on addressing many of the identified barriers (e.g., improving innovation funding processes), there is a desire from industry for greater engagement from Government in co-creating solutions to these problems.
Recommendation 5: Ofgem and BEIS should engage more closely with businesses to address shared challenges and opportunities across the sector.
Conclusions
There is a huge appetite for digitalisation and decarbonisation in the sector, with promising initiatives happening both within and across businesses to build data science capabilities, improve data governance processes and plan for a digital future. To build on and coordinate these initiatives, we believe that the sector would benefit from a roadmap detailing the role of AI in achieving a net zero electricity system.
These changes require evolving regulatory processes (data standards, software development, and innovation funding) which were designed for traditional energy businesses, as well as a cultural shift towards data literacy across the workforce and recognising the value of data. Both are struggling to keep pace with the sector’s ambition.
The CAIC will communicate these workshop outcomes to Ofgem and BEIS and begin enacting these recommendations. We will continue to engage with businesses, researchers and policymakers to enable the adoption of AI to reach a net zero electricity system.
Please get in touch with us if you would like to be involved in any aspect of this work, or if you would like to discuss how we can support your organisation in its data science and AI journey.
The Centre for AI & Climate is a cross-functional facilitator and incubator that aims to connect capabilities across technology, policy, and business to accelerate the application of AI to solve climate challenges. Please visit icaiec.org to find out more