Cluster analysis of residential heat load profiles and the role of technical and household characteristics

Demand-side management methods are expected to play a key role in future energy systems. These methods are making energy demand adaptive to fluctuations inherent to intermittent renewable energy sources. In order to develop feasible demand-side management solutions, a better understanding of the temporality of the energy demand is needed. This paper contributes to this by focusing on the daily load profiles of energy demand for heating of Danish dwellings with heat pumps. Based on hourly recordings from 139 dwellings and employing cluster and regression analysis, the paper explores patterns (typologies) in daily heating load profiles and how these relate to socio-economic and technical characteristics of the included households. The study shows that the load profiles vary according to the external load conditions. Two main clusters were identified for both weekdays and weekends and across load segments; a main cluster with a relatively constant load profile and a minor cluster with a more distinct variation during the day. The difference between the clusters is primarily correlated with building characteristics like floor area, building year and type of space heating distribution system, while the existence of children and teenagers in the household were the only significant socio-economic variables.