Multi-objective optimization for balancing CO2 emissions and thermal comfort in a central heat pump system: A case study of a Danish office building

Buildings consume a large portion of global energy and contribute significantly to CO₂ emissions, with HVAC systems being major energy users. Improving HVAC efficiency is crucial for sustainability and climate goals. Model Predictive Control (MPC) has shown promise in optimizing HVAC operations, enhancing energy savings while maintaining comfort. However, full-scale implementation of MPC in real-world buildings remains limited. This study bridges that gap by developing a high-fidelity energy model of an office building at the University of Southern Denmark to evaluate MPC performance in practice. Unlike traditional methods focusing on terminal components, this research targets the central heat source, a heat pump, and uses MPC to balance CO₂ emissions and thermal comfort. A multi-objective optimization framework is used, weighing CO₂ emissions at 0.6 and thermal comfort at 0.4 when selecting control strategies at each prediction step. Results show that MPC reduced average indoor temperature by 0.74 °C, cut total energy use by 27.96 %, and lowered CO₂ emissions by 26.63 %. Additionally, the optimized heat pump operation aligns energy use with periods of higher renewable energy availability, further reducing dependence on fossil fuels. By shifting the optimization focus to the central heat source and integrating grid-responsive strategies, this study offers a scalable and intelligent solution for decarbonizing HVAC systems. The findings underscore MPC’s potential to enhance building energy performance and contribute to a more sustainable and resilient energy future.