Dr Anthony Woolcock, lead data scientist, Energy Systems Catapult Customers, Energy networks, Energy retail, Innovation, Technology, Opinion, Technology and Innovation Council

Data scientist Dr Anthony Woolcock explains why AI underpins the UK's ability to achieve its decarbonisation ambitions.

Have you noticed how those high-tech ‘homes of the future’, once showcased as a far-flung concept on BBC’s Tomorrow’s World, are beginning to look remarkably reminiscent of homes today?

Artificial intelligence is already in most of our homes, thanks to the near-ubiquitous presence of Alexa and Google Home. But AI has a far more profound role to play in the home. The threat of climate change is causing us to redraw the blueprint of our archaic, largely analogue energy system, and a growing army of innovators is harnessing AI to transform it into one that is smarter, greener and more efficient.

When we talk about AI, what we mean is the use of data and machine learning to understand and predict human behaviour, which can be used to build automated products and services tailored to the unique lives of every user. This ever-advancing technology has the potential to make us all more comfortable whilst maximising efficiency and will be crucial in the mission to reaching net zero emissions.

As we decarbonise energy, its sources will diversify and be less carbon intensive. This may include the use of hydrogen alongside electricity, likely to be partly produced from low carbon sources. Heating and cooling devices are likely to be connected, allowing smarter decisions to be taken about how they are used to maximise comfort and efficiency. What’s more, some of these devices may be able to store energy, such as an electric vehicle, for use at a more opportune time such as when demand is high.

How we pay for energy is also likely to be different in future. Customers may have the opportunity to take advantage of varying energy prices throughout the day. This is likely to result in different tariff structures, which vary depending on energy supply and demand. There is also the possibility of a service-based payment structure. For example, people might purchase hours of warmth in their home rather than kilowatt hours – which few consumers understand.

There are several benefits to customers in this new, more digitalised landscape. AI that describes the usage of heat and related data in an understandable way, enabled by smarter home systems, will help people get a greater understanding of whether they are getting the most out of their heating system.

There are many factors that affect the way homes are heated. These can include the characteristics of each customer (e.g. when we like to schedule our heating) and the characteristics of the home (e.g. it’s insulation the heating system and the time it takes to warm up).

By harnessing data science and AI, we can help users can make more informed decisions about their heating. For example, based on the occupant’s behaviour and stated preferences, a smart system could recommend turning the heating up in the middle of the day. This may go against the grain of their usual thinking, but they would soon realise that this will use less energy and cost less while still improving their overall comfort.

Automated algorithms will also allow for smarter timing decisions. When it’s windy, the grid will have greener and cheaper energy to use. Smart AI decisions to run devices like washing machines or pre-heat the home will help reduce carbon intensity across the grid and can help bring down costs for consumers. The predictive power of AI can help forecast the surplus power and schedule these devices in a smarter way.

Additionally, when people are cooking, making tea or turning on their heating in the evening, there is an opportunity for AI-driven timing decisions to schedule devices to run earlier or later than the predicted peak usage times. This could reduce peak demand and the use of carbon intensive and more expensive energy sources.

AI can also help energy providers shape their tariff structure or energy as a service propositions in a sustainable way that keeps their customers happy. This could include an AI process that feeds directly into business decisions, tailoring each proposition to the customer’s specific needs.

Data and AI have huge potential to make people’s lives better when it comes to home energy. Research conducted in Energy Systems Catapult’s Living Lab has already indicated the potential demand there could be for this kind of experience. 78% of participants in the Catapult’s Heat Plan trial – where participants were given AI-enabled smart home technology to enable them to purchase ‘Warm Hours’ instead of kilowatt hours – said that smart heating controls improved or maintained comfort levels.

For regulators and policy makers, the challenge is to ensure that AI is developed in ways that safeguard consumers’ rights and privacy, while also advancing key policy objectives around affordability, sustainability and security of supply.

It is increasingly clear the new energy landscape can deliver great experiences for consumers. The challenge for energy providers and the wider sector is to understand this new landscape enough to be able to make business strategy decisions that are sustainable and keep customers happy.

This article was commissioned as part of a special series of industry viewpoints on the transformative potential of AI for the utilities sector. This series is being published in the lead up to the next meeting of the Utility Week-Wipro Technology and Innovation Council on 6 November, which will include a unique panel debate on the challenges and opportunities involved for companies seeking to disrupt and add value through using AI. To express you interest in attending this event, please contact Liz Izard at lizizard@fav-house.com