Behavior segmentation of electricity consumption patterns: A cluster analytical approach

作者：

Highlights：

• A large-scale analysis of Slovenian households is presented to find the key factors affecting their energy consumption.

• The time and type of day, season, and holidays are key factors to understand the peak demand and variability of consumers.

• Cluster analysis of the features extracted from these key factors is analyzed to identify the primary consumption profiles.

• The end users of the identified profiles accurately reveal their peak demand in different seasons and type of days.

• An unsupervised anomaly detection method based on the clustering to identify anomalous consumers is presented.

摘要

•A large-scale analysis of Slovenian households is presented to find the key factors affecting their energy consumption.•The time and type of day, season, and holidays are key factors to understand the peak demand and variability of consumers.•Cluster analysis of the features extracted from these key factors is analyzed to identify the primary consumption profiles.•The end users of the identified profiles accurately reveal their peak demand in different seasons and type of days.•An unsupervised anomaly detection method based on the clustering to identify anomalous consumers is presented.

论文关键词：Smart meter,Clustering,Data analysis,Anomaly detection

论文评审过程：Received 3 September 2021, Revised 10 May 2022, Accepted 8 June 2022, Available online 15 June 2022, Version of Record 27 June 2022.

论文官网地址：https://doi.org/10.1016/j.knosys.2022.109236