We’re developing a state-of-the-art, automated forecasting system powered by machine learning to support the UK’s National Grid Electricity Distribution (NGED). NGED is the largest Distribution System Operator in the UK, managing the electricity that keeps the lights on for over 15M customers.
As we transition to a cleaner, renewable energy system, sources of power are becoming smaller and increasingly distributed across the UK. Rooftop solar, electric vehicles and other forms of decentralised sources make it much more difficult for Distribution System Operators to predict electricity demand and ensure everyone's lights stay on.
Working with NGED, we are employing our industry knowledge and leading AI talent to approach this challenge with an innovative dual-track approach. This will enable us to combine cutting edge innovative research with an efficient, practical implementation of the new technology.
The project, aptly named Flexpectation, will result in a robust, operational API-based forecast providing probabilistic predictions for all of NGED’s 1161 primary substations up to 14 days ahead. With this, NGED will be able to manage grid constraints more effectively, activate demand side flexibility more precisely, reducing operational costs, and enhancing the UK’s network stability.