HERS standards plug-load interaction model exploration: incorporating the interaction of independent variables to improve the HERS standards plug-load models

Abstract

The HERS index is the most well recognized measure for residential energy efficiency in the United States. The index provides an asset label in which the energy efficiency of the home’s design can be compared with other similar homes. The asset label is based on building characteristics and standardized occupant behavior. Plug-loads, such as 120 V appliances, refrigerators, televisions, dish washers and clothes washers are notoriously difficult to model. These loads are dependent on the occupant’s specific behavior which cannot be accounted for completely using standardized user behavior profiles. The HERS index accounts for plug-loads by using a regression model that calculates the annual load based on either the size of the home or the number of bedrooms. This study improves this model by not only using the size of the home and the number of bedrooms which are used in the HERS asset label, but also the statistical interaction between these two variables. In addition, this paper introduces new building characteristics such as climates zone, urbanicity and age of home and compares them to the current HERS model. By using additional building characteristics, this study shows that plug-loads can be modeled more accurately than current accepted practices.