National Grid exploring the potential of Artificial Intelligence to optimise renewables

The National Grid has confirmed that it is in the "earliest stages" of discussions exploring the use of Artificial Intelligence (AI), which could potentially maximise the use of renewable energy by predicting peaks in demand across the UK.

The National Grid, which operates and owns the infrastructure that transports electricity across the UK, has seen its ability in balancing and stabilising the grid challenged in recent years as intermittent renewables such as solar and wind have been fed into the energy mix.

While the introduction of renewables into the mix forms a key role in both national and European legislation to decarbonise the grid, concerns have been raised as to the National Grid’s ability to deal with fluctuating wind and solar resources, which can sometimes produce more energy than the system can cope with.

Energy storage and demand response initiatives, whereby businesses either store surplus energy or increase or reduce energy consumption based on demand, are being incorporated by the National Grid, which is now “exploring what opportunities” AI could offer to balance the situation.

The National Grid revealed that it is in discussions with the UK-based AI company DeepMind about introducing new technologies to help balance the grid and improve the use of renewables. DeepMind technology has already been used in Google’s data centres to cut energy by 40%.  

“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,” a spokesperson for 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.”

DeepMind’s machine learning algorithms are able to predict the temperature and pressure outputs within data centres 60 minutes in advance, and the company has suggested that the technology could accurately predict demand patterns in the UK.

The National Grid had to battle with fluctuating demands last summer, when it anticipated that electricity demand would hit a record low of around 35.7GW, despite electricity generation hitting 67.4GW in the build-up.

Power stations had been put on standby by the National Grid since the winter of months of 2014 in an effort to create a supplemental balancing reserve to cope with fluctuating demand. However, the Energy and Climate Intelligence Unit (ECIU) has since found that the scheme, which shut in February this year, was never used – costing £180m in public money over the three-year period.

Although the discussions between National Grid and the AI firm are still in the early stages, DeepMind believes that predictive machine learning could boost the use of renewables by more accurately predicting demand and supply.

A DeepMind spokesperson: “As we’ve said publicly for many months, 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.

“While we’re excited about this idea, we’re at the very earliest stages of exploring a possible partnership.”

$2.4trn opportunity

The discussions took place just days after the World Economic Forum (WEC) released a new report outlining how technological improvements to electricity grids across the globe could increase energy efficiency and generate vast economic returns.

The “Future of Electricity: New Technologies Transforming the Grid Edge” report, released on Friday (10 March), found that new “grid edge” technologies such as smart meters, connected devices and grid sensors could create more than $2.4trn for the utilities industry and society in the next decade.

The economic returns for society will be delivered, the report claims, by placing consumers at the heart of new operating models. Roll-outs of new technology will see consumers produce their own electricity, with options also available to store and consume it when costs fall or sell it back to the grid.

According to the report, the technological advancement of grid systems could even create peer-to-peer decentralised transactions similar to Airbnb and Uber as part of a new sharing economy market.

Matt Mace

Action inspires action. Stay ahead of the curve with sustainability and energy newsletters from edie