Analysis and Clustering of Residential Customers Energy Behavioral Demand Using Smart Meter Data

TLDR

Clustering residential smart meter data offers distribution network operators opportunities to manage low‑voltage networks, identify demand‑response candidates, and improve energy‑profile modeling, but the high stochasticity and irregularity of household demand necessitate detailed analytics to define suitable clustering attributes. The study analyzes customer smart meter data to better understand peak demand and major sources of variability in residential behavior. The authors analyze peak demand and variability, then apply a finite‑mixture‑model clustering to identify ten distinct behavior groups, and use a bootstrap technique to assess clustering reliability. The analysis identifies four key time periods for attribute formation, the clustering yields ten distinct behavior groups, and bootstrap validation demonstrates its reliability—marking the first test of sample robustness in power‑systems literature.

Abstract

Clustering methods are increasingly being applied to residential smart meter data, which provides a number of important opportunities for distribution network operators (DNOs) to manage and plan low-voltage networks. Clustering has a number of potential advantages for DNOs, including the identification of suitable candidates for demand response and the improvement of energy profile modeling. However, due to the high stochasticity and irregularity of household-level demand, detailed analytics are required to define appropriate attributes to cluster. In this paper, we present in-depth analysis of customer smart meter data to better understand the peak demand and major sources of variability in their behavior. We find four key time periods, in which the data should be analyzed, and use this to form relevant attributes for our clustering. We present a finite mixture model-based clustering, where we discover ten distinct behavior groups describing customers based on their demand and their variability. Finally, using an existing bootstrap technique, we show that the clustering is reliable. To the authors' knowledge, this is the first time in the power systems literature that the sample robustness of the clustering has been tested.