Model predictive control of a photovoltaic-powered domestic hot water generation system with electric and thermal energy storage: optimization of operational costs and profitability

Improving energy efficiency and control performance in residential heating systems is vital for cutting greenhouse gas emissions and advancing sustainable construction. This thesis explores the application of Model Predictive Control (MPC) to operate a heat pump system, with the goal of optimizing energy consumption while ensuring indoor thermal comfort amid varying weather and occupancy. MPC is recognized for its capability in addressing complex control problems with constraints, offering more adaptability and sophistication than traditional methods.,A dynamic model representing the heat pump and building's thermal dynamics was crafted in Python and incorporated into the MPC framework. This model forecasts system performance over an upcoming time period, enabling the controller to pinpoint optimal actions by reducing a cost function that balances energy use and comfort. The strategy accounts for predicted disturbances like changes in outdoor temperatures and internal heat profits to enhance control precision and responsiveness.,To assess MPC's effectiveness, it was benchmarked against a standard rule-based control (RBC) strategy, commonly employed in homes due to its straightforwardness. Simulations across various scenarios showed that MPC excels over RBC in energy savings, temperature management, and adaptability. MPC's predictive capability resulted in more seamless and efficient control actions, boosting comfort and curbing energy usage.,The findings underscore MPC’s promise as an intelligent control solution for forthcoming heating systems. It delivers a more adaptable and efficient method, particularly under changing environmental conditions. The thesis also addresses challenges like model precision, computational demands, and practical implementation.