Quartz Solar: AI solar forecasts in Britain's grid control room

⚠ Not yet claimed by the project owner or their organization. This information may not be completely accurate. Is this your project? Claim it.

Open Climate Fix, a nonprofit, built Quartz Solar, which forecasts how much electricity Britain's solar panels will produce. It combines machine learning with live satellite images and weather data, and is now used in the control room of the National Energy System Operator (NESO).

Why it mattersWhen grid operators can't predict solar power well, they keep fossil-fuel plants running as backup. Better forecasts mean less backup, lower costs, and lower emissions.

What they learned

Worked

Large solar forecast errors halved after moving to AI forecasts

From public sources: Open Climate Fix reports that Quartz Solar halved large forecast errors compared with NESO's previous tools, and estimates it saves at least £30 million a year in balancing costs and about 300,000 tonnes of CO2. These figures come from Open Climate Fix, which does not publish its calculation. The tool spent years as a funded innovation project with extensive testing before it went into the control room.

The problem

Solar output swings with the clouds. Forecasts that update only every few hours force the grid to hold expensive, often fossil-fuelled, reserve power.

What the AI does

It looks at live satellite images of clouds and at weather forecasts, then predicts solar generation from minutes to over a day ahead, with a range of likely outcomes rather than a single number.

Sources

This entry was summarized in our own words from these public sources.

Claim this entry Suggest an edit Ask for a correction or removal

Contact the team

We forward your message. We never show anyone's email address.