Prediction of power and energy use in dwellings : Addressing apects of thermal mass and occupant behaviour

Sammanfattning: Households are responsible for approximately 26 % of the annual energy use in the EU. Following the EU-directives regarding energy performance in buildings, international initiatives have been taken in Europe to help countries to define and include guidelines in their own building codes, for example, to establish the concept of zero energy buildings, ZEBs. This concept includes passive building energy-saving technologies, energy-efficient building services systems and renewable energy generation technologies. It is usually very difficult for a building to use zero energy and the concept has therefore been developed to include so-called net-zero energy buildings, or nearly zero energy buildings. These are usually defined as having a net-zero energy use on an annual basis and a nearly-zero energy use if they have a significantly lower use than stipulated in the respective national building codes. Technological advances have resulted in new buildings being very well insulated and, subsequently, using very little energy. However, the focus has now moved towards the use of renewable energy rather than only looking at the amount of energy used. Energy production can be achieved via numerous different arrangements and can be utilized in ways that are dependent on the time of day and the weather. Taking these different aspects into consideration, it can be assumed that the temporal variations regarding production can vary significantly. The heating demand of a building depends on the outdoor climate and the occupants’ behaviour, which leads to an uncertain situation with regard to matching the renewable production and demand, and even more so when the occupants’ behaviour is subject to temporal variations. In addition to the temporal variations, occupant actions or preferences are subject to large stochastic variations within a population. Thus, when designing to meet these challenges, the temporal resolution would have to be higher with regard not only to demand but also to the renewable energy production, in order to provide general benefits as well as covering a larger part of the possible future scenarios.This thesis aims show how the use of household electricity and domestic hot water varies and how these variations impact the energy and power demand of buildings. Additionally, in order to achieve a higher possible level of load matching there is a need to time shift power loads. This is another building operation process that has been investigated. The primary method in both cases has been to use building simulations with large amounts of measurement data for occupant behaviour as input to the simulation models. By randomly inserting different measurement data sets, and running simulations repeatedly, the outcomes were hundreds of annual energy and power demands that varied with the variation of the input.Furthermore, load shifting was investigated by abruptly reducing the heating power supply to buildings. The heat stored in the building envelope and furniture was then used to reduce the effects on the indoor temperature. This thesis examines the temperature drops caused by such power reductions and the various factors that affect the size of the temperature drops, such as the thermal mass and the properties of the building envelope, as well as the stochastic behaviour of the occupants that creates the internal heat load,