Investigating the effect of human behaviour on the energy performance of 3 typical Dutch residential dwellings using sensors and dynamic performance modelling

Abstract

Buildings don’t use energy: People do. Do highly efficient homes really save energy? Counter-intuitively, dwellings that use less energy have low energy efficiency, and those with higher energy consumption have higher efficiency. Moreover, even accurate building energy simulation for multiple buildings could not predict the actual energy consumption. This is termed as the ’energy performance gap’. The regulatory energy performance gap is found to deviate by +34% with an SD of 55% based on 62 case buildings. In certain cases, the energy performance gap due to occupant behaviour was 300%. This is attributed to factors such as - climate, indoor design criteria, building energy service systems, building envelope, building operation maintenance and occupant behaviour. Significant progress in all the stated aspects has been made, except the latter - occupant behaviour.
Occupant behavior is complex, stochastic and multi-disciplinary. Accounting for it is challenging. Typical human interventions like heating/cooling, opening/closing windows, use of sunshades, hot water used, number of electrical appliances, lighting and cooking (gas/electric) play an important role in predicting the actual energy usage. Over-simplification and negligence of these interventions during the design phase is responsible for the energy performance gap. 
This study aims to study the impact of occupant behaviour on the energy consumption of three Dutch residential dwellings with an emphasis on data-informed decisions and predictions. Three candidates - senior couple, young couple and family were selected out of 13 interested candidates. Their houses were observed for 2 weeks from 7th - 21st of March 2019 with the help of sensors and meters. This study proposes a novel, flexible and inexpensive method of data collection with micro-controllers and sensors. The cost of one set-up being as low as € 14. With the help of actual data and the dynamic performance modelling tool - IES(VE) 2018, this study informs the readers about the energy distribution in low and highly energy efficient homes. Comparison of the actual energy consumption with energy standards like EPC and BENG has been made to study their reliability. It also discusses the possibility of sensors and actuators replacing sophisticated energy-siphoning HVAC systems. Lastly, the impact of the three households’ behaviour on their energy consumption has been quantified.