Clustering analysis of residential electricity demand profiles

Abstract

h i g h l i g h t sOptimal k-means clustering finds seasonal groups of residential electricity use. We find that each season has two nominal groups. One group typically uses more expensive electricity than the other. Regression analysis allows for insight as to which homes will be in which cluster. a b s t r a c tLittle is known about variations in electricity use at finely-resolved timescales, or the drivers for those variations. Using measured electricity use data from 103 homes in Austin, TX, this analysis sought to (1) determine the shape of seasonally-resolved residential demand profiles, (2) determine the optimal number of normalized representative residential electricity use profiles within each season, and (3) draw correlations to the different profiles based on survey data from the occupants of the 103 homes. Within each season, homes with similar hourly electricity use patterns were clustered into groups using the kmeans clustering algorithm. Then probit regression was performed to determine if homeowner survey responses could serve as predictors for the clustering results. This analysis found that Austin homes fall into one of two seasonal groups with some homes using more expensive electricity (from a wholesale electricity market perspective) than others. Regression results indicate that variables such as if someone works from home, hours of television watched per week, and education levels have significant correlations with average profile shape, but might vary across seasons. The results herein also indicate that policies such as time-of-use or real-time electricity structures might be more likely to affect lower income households during some high electricity use parts of the year.

Fuzzy clustering and prediction of electricity demand based on household characteristics

“…[20,21] presents studies on the prediction of household information based on smart meter data. In [22,23] consumptions profiles obtained via clustering are correlated to household characteristics in a similar fashion to what is done in this work.…”

“…Based on the work of Rhodes et al [23], the relationship between the demand profiles and survey variables is studied using Probit regression to determine if there are any significant correlations between the survey data and the probability of a consumer being in a certain cluster. The explanatory variables represent the survey data (e.g.…”

Tailor-Made Feedback to Reduce Residential Electricity Consumption: The Effect of Information on Household Lifestyle in Japan

“…Various analysis methodologies have been developed: they have used the regression model [37][38][39][40]; time-series analysis [41][42][43][44][45][46][47][48][49]; and clustering techniques [46][47][48][49][50][51][52][53][54]. However, most analyses have been aimed at short-and medium-term demand forecasting; relatively few analyses have been directed at tailor-made feedback.…”

Electricity Consumption Clustering Using Smart Meter Data

“…Several methods for clustering have been applied and the most prevalent is K-means [5,6] and derivatives such as fuzzy K-Means [7,8] and adaptive K-Means [9]. Further algorithms like hierarchical clustering [10,11], and random effect mixture models [12,13] are also popular.…”