Predictive Control Strategies for aHeat Pump System with PV and Electrical Storage with Various Boundary Conditions

The idea of nearly zero energy building (nZEB) has been emerged from the global trend toward reduction in fossil fuel consumption and green-house gas emission. In spite of different definitions for nZEB, it is vastly known as a grid connected building with highly reduced energy consumption by means of on-site renewable energy production. Accordingly, heat pumps are one of the remarkably efficient heating and cooling technologies considered as the promising components of the nZEBs, specially, when they are associated with the renewable energy systems. However, due to the interaction between heat pump, PV and the grid, the efficiency and the electrical self-consumption of the system strongly depend on the advanced controller strategies such as predictive controllers using price and weather forecast services. Although design of the control algorithms is a very challenging procedure, evaluation of the functionality of these defined algorithms is of paramount importance as well.In this thesis, the robustness of the pre-defined price and weather predictive controllers is studied by testing the performance of a heat pump system for different boundary conditions and compare the results with a reference case study. The base case is a heat pump system for a single-family house with the total heating floor area of 143 m2 and the annual heating demand of 100 kWh/m2.year. The building is located in Norrköping, Sweden and it is assumed that a family of 4 (parents and 2 children) are living in the house. The associated 5.7 kW PV system is mounted on the south oriented roof with the slope of 27˚. Moreover, the system has a storage tank with the total capacity of 180 liter for DHW and a 7.2 kWh battery bank increases the amount of harvested solar energy.The identified methodology of this thesis suggests evaluating the system for the range of variables and boundary conditions, included the climate, thermal properties of the building, orientation and slope of the roof of the building, room set temperature and the occupancy of the building which leads to the various DHW and electricity load profile. Moreover, to study the influences of the controllers on the system, a group of performance indicators are defined. PV self-consumption, solar fraction, final energy use, annual net cost of electricity and seasonal performance factor of the system are considered as the key figures of this study. Next, it aims to do the sensitivity analysis of the system with and without controllers under various boundary conditions. For this purpose, the TRNEdit, as one of the TRNSYS tools which is extended purposely to edit TRNSYS files and manage the parametric simulation studies, is used. Finally, the results from the parametric studies of the system is evaluated to examine the robustness of the controllers.Consequently, the acquired results from the sensitivity analysis of the system with the introduced predictive controllers proves that the performance of the system successfully promotes when it utilizes predictive strategies of PV generation and electricity price. However, the suggested control algorithms need to be slightly modified in order to achieve better results when they operate simultaneously