Activity-based modeling of domestic energy demand with high time resolution - A case study on Gothenburg

Abstract

In domestic buildings, a great potential to reduce energy consumption exists. To achieve this reduction, a more detailed understanding of the energy demand is required. Currently, there is sufficient information on an aggregated level. Nevertheless, further knowledge about the timing of the energy use is needed. Previous attempts to gain that knowledge are mostly based on conclusions from measurements. However, in this study a bottom up, activity-based modeling approach is chosen. Hence, knowledge can be increased on a more detailed level and the generation of synthetic load curves can be facilitated. The derived model focuses on residential buildings and is applied in a case study on Gothenburg, Sweden. Existing bottom-up models are mostly focused on specific dwellings and require a large amount of detailed input data. In contrast, the goal of the present report is a simple model with as few input parameters as possible. Thus, it can be quickly applied and transferred to other cities in Europe to get a first estimation on the composition of the energy demand. The practicality of such a model and its limitations are assessed by the present study. Furthermore, areas with low data availability and uncertain data should be identified. To gain more understanding of the timing of energy use, four end-uses are modeled. Those are electrical appliances, lighting, hot water and heated rooms. The major focus lies on the consumer side. A large part of the work is the collection of reliable input data. Important input data sets are the time-use surveys and data about the weather (temperature, solar irradiation and natural illuminance). The modeling of the hot water and the electrical appliances is heavily based on the time-use surveys. For the lighting those are combined with the natural illuminance. To model the space heating demand, the energy balance over the building envelope is calculated and combined with information on the external temperature, the solar irradiation and the building stock. The developed model can be used to predict the general shape of the demand curve in a residential area. However, the levels of the peak are rather uncertain. Nevertheless, in combination with measurements of the aggregated demand the model can be a useful tool to assess the current situation in an existing city. The input data needed to model the demand curve’s shape is extractable from established, official sources and data bases. That data II ABSTRACT includes weather data and time use data. Since that is available for other European countries, the model could be transferred to different cities. From the present study, it can be concluded that the time-use surveys are a promising tool for energy load curve modeling. Moreover, it could be detected how and where those surveys can be adjusted for that purpose.