Artificial intelligence (AI) in district heating

In collaboration with Stadtwerke Norderstedt and the start-up RAUSCH Technology GmbH, dena has developed an implementation guide and recommendations for the use of artificial intelligence (AI) in district heating. With the support of numerous experts from the heating sector, ten specific use cases were identified and evaluated. One of the use cases resulted in the prototypical implementation of an AI-based heat load forecast at Stadtwerke Norderstedt. 

The optimisation of heat generator operation and the optimisation of customer systems proved to be particularly promising use cases for AI. The implementation of the AI-based heat load forecast led to a reduction in forecasting errors of up to 25 per cent. Based on this, optimisations were set up so that the most cost-effective generation mix could be determined at all times.

For the implementation of AI-based data analysis projects, recommendations for the comprehensive development of a solid database, targeted data collection strategies and the expected challenges for AI solution providers and district heating supply companies were derived and set out in a comprehensive implementation guide.