TY - THES A1 - Szejwian, Jakub Jan T1 - Behavior analysis of a model predictive controller in a building heating system BT - Impact of cost function adjustments N2 - The increasing global significance of green energy utilization and the reduction of fossil fuel dependency require innovative solutions for energy management in building systems. This thesis investigates the application of Model Predictive Control (MPC) for optimizing heating systems, particularly through the enhancement of cost function design. By utilizing MPC, it is possible to improve the operational efficiency of heat pumps (HPs), resulting in substantial cost savings and energy conservation. The research introduces a test-building model developed at the Institute for New Energy Systems in Ingolstadt, Germany. This model simulates a two-story single-family house equipped with advanced heating systems, including an air-to-water HP, thermal buffer storage (TBS), and heat distribution systems, such as ceiling heating and a thermally activated building system (TABS). The MPC framework utilizes detailed state and control variables alongside disturbance models to predict and optimize the system's performance. Key aspects of this thesis include the development of cost functions that prioritize the efficient utilization of the designed HP by optimizing their operating times relative to ambient temperature conditions. These cost functions aim to maximize the Coefficient of Performance (COP) of the HP, minimize rapid power changes to extend the lifespan of HP’s components, and maintain the conditions within the building for comfortable living standards. Simulations were conducted to evaluate the effectiveness of these cost functions. The results demonstrate that the proposed MPC approach can maintain thermal comfort within the building. The findings highlight the potential of MPC to enhance the sustainability and efficiency of building heating systems, providing valuable insights for future research and practical implementations in energy management. The thesis concludes with a suggestion that further research of weighting variables and a comparative analysis with traditional rule-based controllers could provide deeper insights into MPC's benefits. Overall, this study contributes to advancing sustainable building technologies and offers a robust framework for optimizing energy usage in residential heating systems. Y1 - 2024 UR - http://nbn-resolving.de/urn/resolver.pl?urn:nbn:de:bvb:573-51336 CY - Ingolstadt ER -