Abstract:The large amount of data collected by smart meters is a valuable resource that can be used to better understand consumer behavior and optimize electricity consumption in cities. This paper presents an unsupervised classification approach for extracting typical consumption patterns from data generated by smart electric meters. The proposed approach is based on a constrained Gaussian mixture model whose parameters vary according to the day type (weekday, Saturday or Sunday). The proposed methodology is applied to a real dataset of Irish households collected by smart meters over one year. For each cluster, the model provides three consumption profiles that depend on the day type. In the first instance, the model is applied on the electricity consumption of users during one month to extract groups of consumers who exhibit similar consumption behaviors. The clustering results are then crossed with contextual variables available for the households to show the close links between electricity consumption and household socio-economic characteristics. At the second instance, the evolution of the consumer behavior from one month to another is assessed through variations of cluster sizes over time. The results show that the consumer behavior evolves over time depending on the contextual variables such as temperature fluctuations and calendar events.

No abstract

This paper presents the application of clustering algorithms to daily energy consumption curves of buildings. Our aim is to identify a reduced set of consumption patterns for a tertiary building during one year. These patterns depend on the temperature throughout the year as well as the type of the day (working day, work-free day and school holidays). Two clustering approaches are used independently, namely the K-means algorithm and the Expectation-Maximization algorithm based on Gaussian Mixture Model (EM-GMM). The clustering results obtained with the two algorithms are analyzed and compared. This study represents the first step towards the development of a prediction model for energy consumption.