Electricity consumption clustering using smart meter data

Alexander Tureczek*, Per Sieverts Nielsen, Henrik Madsen

*Corresponding author for this work

Research output: Contribution to journalJournal articleResearchpeer-review

2483 Downloads (Pure)

Abstract

Electricity smart meter consumption data is enabling utilities to analyze consumption information at unprecedented granularity. Much focus has been directed towards consumption clustering for diversifying tariffs; through modern clustering methods, cluster analyses have been performed. However, the clusters developed exhibit a large variation with resulting shadow clusters, making it impossible to truly identify the individual clusters. Using clearly defined dwelling types, this paper will present methods to improve clustering by harvesting inherent structure from the smart meter data. This paper clusters domestic electricity consumption using smart meter data from the Danish city of Esbjerg. Methods from time series analysis and wavelets are applied to enable the K-Means clustering method to account for autocorrelation in data and thereby improve the clustering performance. The results show the importance of data knowledge and we identify sub-clusters of consumption within the dwelling types and enable K-Means to produce satisfactory clustering by accounting for a temporal component. Furthermore our study shows that careful preprocessing of the data to account for intrinsic structure enables better clustering performance by the K-Means method.

Original languageEnglish
Article number859
JournalEnergies
Volume11
Issue number4
ISSN1996-1073
DOIs
Publication statusPublished - 1 Apr 2018

Keywords

  • Autocorrelation
  • Data analysis
  • Electricity consumption clustering
  • K-Means
  • Smart meter analysis

Cite this

@article{e8e297dc7f27401eae43e3dadbfaedf3,
title = "Electricity consumption clustering using smart meter data",
abstract = "Electricity smart meter consumption data is enabling utilities to analyze consumption information at unprecedented granularity. Much focus has been directed towards consumption clustering for diversifying tariffs; through modern clustering methods, cluster analyses have been performed. However, the clusters developed exhibit a large variation with resulting shadow clusters, making it impossible to truly identify the individual clusters. Using clearly defined dwelling types, this paper will present methods to improve clustering by harvesting inherent structure from the smart meter data. This paper clusters domestic electricity consumption using smart meter data from the Danish city of Esbjerg. Methods from time series analysis and wavelets are applied to enable the K-Means clustering method to account for autocorrelation in data and thereby improve the clustering performance. The results show the importance of data knowledge and we identify sub-clusters of consumption within the dwelling types and enable K-Means to produce satisfactory clustering by accounting for a temporal component. Furthermore our study shows that careful preprocessing of the data to account for intrinsic structure enables better clustering performance by the K-Means method.",
keywords = "Autocorrelation, Data analysis, Electricity consumption clustering, K-Means, Smart meter analysis",
author = "Alexander Tureczek and Nielsen, {Per Sieverts} and Henrik Madsen",
year = "2018",
month = "4",
day = "1",
doi = "10.3390/en11040859",
language = "English",
volume = "11",
journal = "Energies",
issn = "1996-1073",
publisher = "M D P I AG",
number = "4",

}

Electricity consumption clustering using smart meter data. / Tureczek, Alexander; Nielsen, Per Sieverts; Madsen, Henrik.

In: Energies, Vol. 11, No. 4, 859, 01.04.2018.

Research output: Contribution to journalJournal articleResearchpeer-review

TY - JOUR

T1 - Electricity consumption clustering using smart meter data

AU - Tureczek, Alexander

AU - Nielsen, Per Sieverts

AU - Madsen, Henrik

PY - 2018/4/1

Y1 - 2018/4/1

N2 - Electricity smart meter consumption data is enabling utilities to analyze consumption information at unprecedented granularity. Much focus has been directed towards consumption clustering for diversifying tariffs; through modern clustering methods, cluster analyses have been performed. However, the clusters developed exhibit a large variation with resulting shadow clusters, making it impossible to truly identify the individual clusters. Using clearly defined dwelling types, this paper will present methods to improve clustering by harvesting inherent structure from the smart meter data. This paper clusters domestic electricity consumption using smart meter data from the Danish city of Esbjerg. Methods from time series analysis and wavelets are applied to enable the K-Means clustering method to account for autocorrelation in data and thereby improve the clustering performance. The results show the importance of data knowledge and we identify sub-clusters of consumption within the dwelling types and enable K-Means to produce satisfactory clustering by accounting for a temporal component. Furthermore our study shows that careful preprocessing of the data to account for intrinsic structure enables better clustering performance by the K-Means method.

AB - Electricity smart meter consumption data is enabling utilities to analyze consumption information at unprecedented granularity. Much focus has been directed towards consumption clustering for diversifying tariffs; through modern clustering methods, cluster analyses have been performed. However, the clusters developed exhibit a large variation with resulting shadow clusters, making it impossible to truly identify the individual clusters. Using clearly defined dwelling types, this paper will present methods to improve clustering by harvesting inherent structure from the smart meter data. This paper clusters domestic electricity consumption using smart meter data from the Danish city of Esbjerg. Methods from time series analysis and wavelets are applied to enable the K-Means clustering method to account for autocorrelation in data and thereby improve the clustering performance. The results show the importance of data knowledge and we identify sub-clusters of consumption within the dwelling types and enable K-Means to produce satisfactory clustering by accounting for a temporal component. Furthermore our study shows that careful preprocessing of the data to account for intrinsic structure enables better clustering performance by the K-Means method.

KW - Autocorrelation

KW - Data analysis

KW - Electricity consumption clustering

KW - K-Means

KW - Smart meter analysis

UR - http://www.scopus.com/inward/record.url?scp=85045410482&partnerID=8YFLogxK

U2 - 10.3390/en11040859

DO - 10.3390/en11040859

M3 - Journal article

VL - 11

JO - Energies

JF - Energies

SN - 1996-1073

IS - 4

M1 - 859

ER -