ODIN - Optimisation and Diagnostics for Innovative Networks

Engineers at high-voltage direct current halls currently spend hours manually sifting through robot monitoring data, looking for signs of trouble. This labour-intensive process cannot easily spot emerging trends or compare current readings against normal operating conditions. As the UK pushes to expand its high-voltage direct current network—a critical part of transmitting renewable energy over long distances—the sheer volume of monitoring data will overwhelm manual methods. The ODIN project replaces that human slog with automated machine learning and artificial intelligence tools that continuously interpret the data stream. If successful, the system will detect subtle changes in asset behaviour long before they become failures, improving the reliability and resilience of the electricity grid. That matters because high-voltage direct current links are the backbone of the net-zero energy transition, connecting offshore wind farms and intercontinental power cables to where the electricity is needed. The research does not aim to discover new physics or materials; it applies existing analytics to a specific industrial bottleneck. The payoff is operational: fewer outages, lower maintenance costs, and a grid that can handle more renewable power without breaking down.