Inferring daily routines from electricity meter data

• Daily routines are inferred from electricity meter data measured every 15 min.
• The activity index is proposed to identify daily human activity from interval meter data.
• The variation index is proposed to extract daily routines from interval meter data.
• Results show that the indexes detect activity and daily routines from 241 anonymous meters.
• Statistical analysis of the activity and daily routines extracted from the data.

This paper studies how human activity affects energy consumption and how such activity can be leveraged from electricity meter data. Using 15-minute interval data and applying the methods introduced in this paper, it is possible to identify in which days there was human activity and extract daily routines from such data. This type of insight into the way people consume energy helps understanding energy use at a deeper level, and realise where significant energy savings can be made.