Accessibility
Language
More precise heat forecasts thanks to new technology
New forecasting technology improves efficiency and sustainability in district heating supply
21.01.2025
Source: E & M powernews
Researchers at the Fraunhofer Institute IOSB-AST want to improve the quality of day-ahead forecasts for district heating and are relying on a new technology to do so.
A major challenge for district heating operators is often forecasting the required heat output in the grids. According to researchers at the Fraunhofer Institute of Optronics, System Technologies and Image Exploitation, Applied Systems Technology (IOSB-AST), often only simple feed-in forecasts are used for the power plants, which do not adequately capture the actual heat demand and do not take sufficient account of the flexibility of the heating network. The IOSB-AST announced on January 21 that they want to use new forecasting approaches to take into account the complex dynamics in the grids in future and thus increase the quality of the forecasts.
On behalf of "TEAG Thüringer Energie AG", researchers at Fraunhofer IOSB-AST say they have now made significant progress in district heating forecasting. At the heart of this is smart data pre-processing in combination with artificial intelligence methods based on the Long Short Term Memory (LSTM) method. The consumption behavior within the heating network is not measured directly, but rather learned from available metadata, such as measured values at individual nodes. "The storage potential of the heating network can then be optimally used for predictive operation of the generation units, for example to improve day-ahead electricity marketing or reduce the use of resources," writes the research institute.
Such precise forecasts are becoming more important: on the one hand, existing heating networks must achieve a share of 80 percent renewable energy and unavoidable off-take by 2040 - 65 percent is already mandatory for new heating networks from March of this year. Secondly, heating networks are becoming more complex with each new component, such as decentralized large heat pumps or CHP plants. This makes it all the more important to consider the cost and revenue side, in particular the marketing of electricity, believe the researchers at Fraunhofer IOSB-AST.
Author: Heidi Roider
