K-means clustering of electricity consumers using time-domain features from smart meter data

Abstract

Abstract Smart meter stores electricity consumption data of every consumer in the smart grid system. A better understanding of consumption behaviors and an effective consumer classification based on the similarity of these behaviors can be helpful for flexible demand management and effective energy control. In this paper, we propose an implementation of unsupervised classification for categorizing consumers based on the similarity of their typical electricity consumption behaviors. The main goal is to group similar observations together in order to easily look at the dataset. Hence, we go through pattern identification in households’ consumption with the K-means clustering algorithm. K-means clusters consumption behaviors based on extracted temporal features into k groups. The result from the algorithm helps power suppliers to understand power consumers’ better and helps them make better informed decision based on the information available to them. The dataset used in this paper is a real data from the London Data Store energy consumption readings for a sample of 5567 London Households that took part in the UK Power Networks Led Low Carbon London project between November 2011 and February 2014 available at: https://data.london.gov.uk/dataset/smartmeter-energy-use-data-in-london-households .

Keywords

• Clustering
• Feature extraction
• K-means
• Smart grid
• Smart meter