Alphabet’s London-based AI outfit DeepMind and the National Grid are in early-stage talks to reduce the UK’s power usage purely through neural networks and machine learning—no new infrastructure required.
Demis Hassabis, co-founder and CEO of DeepMind (and lead programmer on Peter Molyneux’s Theme Park), hopes that the UK’s energy usage could be reduced by as much as 10 percent, just through AI-led optimisation. The UK generated around 330 terrawatt-hours (TWh) of energy in 2014, at a cost of tens of billions of pounds—so a 10 percent reduction could be pretty significant, both in terms of money spent and carbon dioxide produced.
The National Grid, owned by a publicly traded company of the same name, owns and operates the UK’s power transmission network—that is, the country’s power lines and major substations. The sources of energy—power stations, hydro plants, wind turbines, and a smattering of solar panels—are owned by other big companies (primarily EDF and E.On).
Importantly, though, it is the National Grid’s job to balance supply and demand across the network, so that the AC frequency that arrives at your house is always within ±1% of 50Hz. Energy demands are usually quite predictable, in that they closely align with standard human behaviour (waking and sleeping hours) and the weather. Energy supply, however, is much less reliable, especially as the UK adds more wind and solar power to the mix.
While the UK has about 13 gigawatts of installed wind power capacity—the nation’s average power draw is only about 35 gigawatts, incidentally—a lack of wind can cause major issues. Back in November 2015, the last time we had a major power shortfall in the UK, all those wind turbines only produced about 400 megawatts. (You should read that story if you want more information about how the National Grid works, and how it uses short-term reserves to balance supply and demand.)
Ingesting data, predicting trends, and suggesting solutions is almost perfectly suited to DeepMind’s neural network expertise. While the National Grid is surely aware of some potential optimisations, a more rigorous investigation by a DeepMind AI may uncover solutions that the grid’s human operators have never considered. One thing’s for certain: a system as large as the UK grid has millions of inefficiencies. The biggest losses come from long-distance power transmission and voltage transformers, but it all adds up.
“There’s huge potential for predictive machine learning technology to help energy systems reduce their environmental impact. One really interesting possibility is whether we could help the National Grid maximise the use of renewables through using machine learning to predict peaks in demand and supply,” DeepMind told the Financial Times.
The National Grid said: “We are in the very early stages of looking at the potential of working with DeepMind and exploring what opportunities they could offer for us. We are always excited to look at how the latest advances in technology can bring improvements in our performance, ensure we are making the best use of renewable energy, and help save money for bill payers.”
Last year DeepMind performed a similar analysis on Google’s data centres, apparently netting a 15 percent reduction in electricity usage. DeepMind trained a neural network to more accurately predict future cooling requirements, in turn reducing the power usage of the cooling system by 40 percent. “Because that’s worked so well we’re obviously expanding that capability around Google, but we’d like to look at doing it at National Grid-scale,” Hassabis said to the FT.
“We think there’s no reason why you can’t think of a whole national grid of a country in the same way as you can the data centres.”
Did you know thatTechnica now has a weekly newsletter? It contains all of the week’s top stories, plus info about upcoming meetups and other events. Sign up now.