Dalarna University's logo and link to the university's website

Uncertainties in occupant behavior lead to discrepancies between the actual energy demand and the predicted results. This paper aims to explore the influence of the variation in occupancy patterns on domestic energy demand prediction. The smart meter data shows the significant differences in occupancy patterns, which is resulted from the impact of the floor area, the number of occupants, the occupants’ habit of using appliance. Single-zone simulation with internal load profiles of each household suggests that the variation in occupancy patterns would result in the significant variation on domestic energy demand prediction. Also, single-zone simulation with different levels of the internal load schedules shows the simulation accuracy is not obvious by using the different level of profiles. It reveals that the use of different levels of internal load schedules depends on how much information could be obtained for simulation analysis, as well as the purpose and scale of the analysis. Another important finding is that nearly all simulation results give higher energy use than the actual demands, reflecting the pre-bound effect. © 2017 Building Simulation Conference Proceedings. All rights reserved.