Digital Transformation in District Heating Systems: A Scoping Review

The application of digital technologies provides the potential to optimize district heating system operations. However, there are also several challenges to realizing the full potential of digitalization in district heating systems. Therefore, this paper conducts a scoping review of thirty articles with thematic analysis and qualitative synthesis. The analysis result reveals that six district heating segments have been discussed in the literature: district heating network, buildings, district heating substations, combined heat and power, renewable energy sources, and district heating underground pipes. District heating networks and buildings are the two main focuses. However, the integration of thermal storage into the district heating system and sector coupling between electricity grids and district heating networks are rarely discussed. Furthermore, Artificial Intelligence techniques, especially Machine Learning, are popularly applied in the literature for heat load/demand forecasting, operation schedule prediction, dataset creation, prediction model improvement, fault detection, evaluation and optimization, and digital twin development. Heat load/demand forecasting is the main focus and different ML methods have been applied, e.g., Regression algorithms, Decision Tree Algorithms, Artificial Neural Networks, and Support vector machines.