The residential sector is responsible for 26% of the European final energy consumption. As such, it represents an opportunity to meet the 2030 European goals for mitigation of global warming and energy use. The Energy Performance of Building Directives (EPBD) made a step in that direction by defining, among others, a framework for energy performance standards. Moreover, buildings envelopes and installations such as HVAC systems have been greatly improved in the past years. However, discrepancies between calculated and real energy consumptions have been observed. These are due to a major parameter that have been standardised for a long time: user behaviour. Based on Time Use Survey databases, behavioural models describing occupancy, activities and so on were developed in order to get a better grip on user behaviour. While researchers now have a better insight into how people are using their houses, it is important to remember that occupants still don’t. Therefore, the next step in reducing residential energy consumption is to provide them with appropriate feedback so that they can improve their behaviour with respect to energy efficiency.

The aim of this research is to create contextual feedback mechanisms that will allow householders to understand and reduce their energy consumption. A preliminary state of the art showed that a successful feedback tool present several key-features : it should be adapted to the context (provide the relevant feedback type on the appropriate time scale), use normative messages (using peer comparison), and be applicable in an opt-out program. Step by step, a plug-load consumption feedback tool, compatible with these three observations, will be developed.

The project methodology has been established in four work packages. First, several studies will be monitored within the Smart City Block project. Gas and disaggregated electricity consumption will be recorded in five different households (WP1). The data collected in WP1 will be used to create and evaluate a first prototype of a plug-load consumption feedback tool. For each monitored household, a synthetic peer group will be created using a probabilistic behavioural model. Personalized contextual feedback will then be provided to each household through a smartphone app highlighting the differences between the real household, the average peers and the energy-efficient peers (WP2). Then, the work of WP2 will be adapted in order to build an academic prototype of energy consumption feedback tool. This will be done by performing dynamic building simulations in order to find building-related energy use for the peer group and by implementing feedback about those variables in the plug-load consumption feedback tool (WP3). Finally, the plug-load consumption feedback tool will be finalized in WP4. This include refining of the model and large scale testing, in collaboration with a third party energy service company.

This research is of use not only to energy service companies, but also to every Belgian household that could reduce their energy consumption and energy bills if such a feedback program was implemented at the national scale in association with energy supply companies.