Insights into a Field Test of a Predictively Controlled Air- Source Heat Pump in Single-Family Homes using non-linear optimization
Heat pumps are a key technology for decarbonizing residential heating, with air-source heat pumps being the most common choice in German single-family homes. However, their efficiency decreases during winter when heating demand is highest. Predictive control approaches such as Model Predictive Control (MPC) can mitigate this limitation by shifting operation to periods with more favourable ambient conditions while maintaining thermal comfort. This paper presents the design, implementation, and field evaluation of a predictively controlled air-source heat pump system that utilizes the building mass as thermal storage. The controller was tested in a single-family house equipped with ceiling heating, concrete core activation, and a thermal buffer storage. Simplified models were developed to ensure computational feasibility. A hierarchical control structure combining long-term (7 days) and short-term (16 hours) MPC was implemented. First evaluations of the field measurements show an improvement of 4-27% in the heat pump performance factor by shifting operation to higher ambient temperatures. However, the use of building thermal mass led to increased total electrical consumption, highlighting the drawback of thermal activation and the need for careful implementation of such control algorithms. Future work will extend the evaluation to a full heating season and further refine the controller design.