Abstract

In this study, a new data mining-based methodology is developed to evaluate energy-related behavior of occupants in residential buildings.
In sections 3.1 and 3.2 Occupant Activity Indicator (OAI) and Residential Energy Intensity Indicator (REII) are introduced as two new definitions which are used in this study. The proposed methodology to evaluate the energy-related behavior of the buildings’ residents is based on the difference between the target REII and actual REII. The dissimilarity, which is found between the target and the actual REII, can be used to calculate the potential energy wastage/saving by occupants in different zones and different times in the building.
The practicality of the proposed data mining framework is tested by applying it to a one-year dataset collected in a three-bedroom apartment in Lyon, France. The methodology applied to all zones of the apartment to evaluate the occupants’ energy-related behavior in different zones. As a result, the time and location for potential energy savings by occupants is identified.
The obtained results show that occupants need to be more cautious about their energy consumption in zones 2 and 3 of the apartment. Moreover, the possible energy-wastage behavior in zones 1 and 4 is less than zones 2 and 3, even though the contribution of zone 4 to the energy consumption is significantly higher than the other zones. Besides, by the developed methodology location and time for the best and the worst energy-related behavior by the building's occupants are defined. Furthermore, the variations of occupants' energy-related behavior in the apartment, are identified by time of day, day of week, and months.
Employing the proposed methodology is beneficial for buildings’ occupants to raise their awareness regarding energy consumption. Also, it gives the decision-makers a practical insight into the system behavior, enabling them to create incentives/charges for residential buildings’ inhabitants to modify their energy-related behavior.