As the push for greener energy solutions intensifies, making district heating
(DH) systems more efficient becomes increasingly important. As many existing DH networks still rely on carbon-heavy energy sources, which underscores the need for smarter energy management. This dissertation focuses on how smart heat meters (SHMs) installed in buildings can help optimize energy use and detect issues in DH systems. Key insights include how machine learning can predict energy needs for heating and hot water, the role of occupant behavior in energy usage, and new methods for early fault detection and diagnosis using SHM data. These findings provide practical solutions for DH operators to improve performance, lower costs, and reduce carbon emissions.

For a copy of the thesis, please email inst.build.phd@build.aau.dk.