Probabilistic commercial load profiles at different climate zones

This paper presents a numerical representation for commercial load profiles based on different climate zones. The load profiles of 16 commercial buildings located in 935 cities representing 50 States in the United States (U.S.) are clustered using k-means considering the geographic coordinates and the time zones. The geographic coordinate works as a local criterion to assign a climate zone for cities within the state, the time zone acts as a regional criterion to group cities with the same climate zone in different states based on their time zones. The clusters are evaluated using both external and internal validity indices. A total of 16 annual load profiles are used as representative for 16 different climate zones for each commercial building in this paper. Unlike the prevalent illustration for the commercial load profiles in the literature using graph representation, the obtained profiles in this study are numerically presented. This paper contributes to the literature by proposing a numerical representation for commercial load profiles at 16 climate zones, which are in turn can be used widely in smart grid application.