Data science

UK to Leverage Cloud-Predicting AI to Anticipate Solar Energy Supply

Solar and wind plants provide clean, affordable electricity on large scales, but they have one major bugbear: intermittency. Essentially, intermittency refers to the fact that wind and solar energy are both, to some degree, unpredictable and uncontrollable – and, in the context of a power grid, that means that more controllable generation plants – like fossil fuel plants – have to be ready to fill in the gaps at a moment’s notice so that customers experience no service interruptions. Now, the UK – which relies on renewables for a plurality of its electricity supply – is turning to AI to reduce the functional intermittency of solar power.

In general, solar energy is less intermittent than wind, owing to the predictability of daytime, nighttime, solar intensity and the angle of the sun relative to a location throughout the day. Clouds, however, throw a wrench into the works, chaotically disrupting the supply of solar energy to solar panels with little warning (large-scale climate and weather forecasting models are, broadly speaking, unable to resolve at the level of individual clouds). Complicating matters, energy operators, while aware of large-scale solar facilities, are often unaware of the exact geographic siting of solar panels on households or businesses. The combination of difficult-to-predict clouds and missing location information for many solar panels means that the operators don’t know when clouds are covering those solar panels – and, as a result of that uncertainty, the grid requires a larger buffer of other energy sources to account for the difference.

Enter Open Climate Fix, a nonprofit product lab “totally focused on reducing greenhouse gas emissions as rapidly as possible.” The nonprofit has teamed up with the UK’s National Grid Electricity Service Operator (ESO) to deliver “nowcasting” using machine learning. In essence, the model will train a machine learning model to understand from satellite images how clouds are moving and where sunlight will fall. A separate project, meanwhile, is working to map the UK’s solar panels.

“Accurate forecasts for weather-dependent generation like solar and wind are vital for us in operating a low carbon electricity system. The more confidence we have in our forecasts, the less we’ll have to cover for uncertainty by keeping traditional, more controllable fossil fuel plants ticking over,” said Carolina Tortora, head of innovation strategy and digital transformation at National Grid ESO. “We’re increasingly using machine-learning to boost our control room’s forecasts, and this latest nowcasting project with Open Climate Fix – whose work could have real impact for grid operators around the world – will bring another significant step forward in our capability and on our path to a zero carbon grid.”

This work builds on prior machine learning efforts from the ESO, which have already resulted in a 33% improvement in solar forecast accuracy in recent years.

Back to top button