UK Power Networks, the UK's largest electricity distribution network operator, is launching a new project that uses artificial intelligence (AI) to estimate solar power generated from rooftop installations that are not directly monitored by grid systems.

The initiative, titled 'AI for Visibility and Forecasting of Renewable Generation', is being developed in partnership with Open Climate Fix (OCF). It aims to improve the accuracy of solar power forecasting by making behind-the-meter rooftop solar more visible to grid operators.

UK Power Networks owns and maintains the electricity cables and substations that deliver power to over 8 million homes and businesses across London, the South East, and East of England.

According to Luca Grella, Head of Innovation at UK Power Networks, the AI model draws on data from the company's Open Data Portal, satellite weather data, and substation measurements. The tool does not rely on physical imagery, but instead uses historical and real-time data to estimate unmetered solar output feeding into the grid.

The project addresses a long-standing challenge in energy forecasting: understanding how much energy is being generated by residential solar systems not connected to formal metering systems. This "invisible" generation makes grid balancing more complex, particularly on sunny days when solar output spikes.

Dan Travers, Co-founder of Open Climate Fix, said the project shows how AI can create real-world tools to help accelerate the UK's energy transition.

With rooftop solar expected to significantly grow--total UK solar capacity is projected to reach 90GW by 2035--this AI-powered model is seen as a tool to unlock grid capacity for more low-carbon technologies.

Source: smart-energy.com