Energinet AI-Powered Demand Forecasting Delivers €47M Annual Savings

Title

Energinet AI-Powered Demand Forecasting Delivers €47M Annual Savings

Content

In 2023, Danish transmission system operator Energinet launched an AI-powered demand forecasting and grid optimization pilot across its 400 kV and 150 kV networks, covering Denmark's entire electricity system. By late 2025, the pilot had demonstrated a 28% improvement in day-ahead load prediction accuracy, a 14% reduction in redispatch costs, and an estimated annual saving of €47 million in balancing market expenditures. The programme has since transitioned from pilot to permanent operational status and is being studied by transmission operators across Europe as a reference implementation for AI-enabled grid management. The project was initiated under Energinet's Innovation Strategy 2023-2028, with a total budget of €12 million over three years. Funding came from Energinet's regulated innovation allowance approved by the Danish Utility Regulator and a €3.2 million grant from the EU's Horizon Europe programme. The project was structured as a partnership between Energinet's system operations division, the Technical University of Denmark (DTU) Department of Wind and Energy Systems, Finnish AI company Elisa IndustrIQ for the machine learning platform, and Siemens Energy for integration with Energinet's existing SCADA and energy management systems. The AI system operates on three interconnected layers. Short-term load forecasting (0 to 4 hours ahead) uses a transformer-based neural network architecture trained on 15-minute interval data from 4,200 grid measurement points, weather station observations, and real-time market price feeds. The model processes approximately 2.8 million data points per forecast cycle, generating probabilistic load forecasts with confidence intervals. Day-ahead forecasting incorporates weather ensemble predictions, calendar effects, and historical patterns, while week-ahead and month-ahead layers optimize unit commitment and maintenance scheduling. Wind generation forecasting accuracy improved MAPE from 11.3% to 7.8% (31% reduction), and peak demand forecast error reduced from ±380MW to ±245MW (36% reduction). System frequency deviations exceeding ±100 mHz decreased by 22%. The combined annual savings include €31M balancing cost reductions, €11M redispatch cost reductions, and €5M reduced renewable curtailment, with ROI 3.9x in first year and payback 14 months.