Comparison between regression models of actual and theoretical heating energy use

Panel: 8. Monitoring and evaluation: building confidence and enhancing practices

This is a peer-reviewed paper.

Authors:
Daša Majcen, University of Geneva, Switzerland
Laure Itard, OTB Research Institute
, The Netherlands
Henk Visscher, OTB Research Institute, The Netherlands

Abstract

Dwellings represent a big potential for future energy savings. An European Performance of Buildings Directive (EPBD) has set the guidelines for dwelling performance certification, called energy label already in 2002. Label certificates in The Netherlands were issued in 2007 to encourage the transformation of the dwelling stock towards a lower energy consumption. The certificate provides each labelled dwelling with a theoretical energy consumption for heating, which is in most cases not a good estimation of the dwellings consumption in practice. As a consequence, using this consumption for payback time calculation, policy targets and similar leads to wrong conclusions. In this paper, we analyse the consumptions of dwellings from two datasets in order to find out which factors are responsible for the variation in actual and theoretical heating energy consumption. The analysis shows that whereas dwelling characteristics alone explain 64,3% variation in theoretical gas consumption, all available variables (including dwelling and household characteristics, occupant behaviour and comfort) can only explain 23,8% of variation, out of which more than half is accounted for by occupant characteristics.