Advanced search
1 file | 2.02 MB Add to list

Identification of energy use time patterns of occupied dwellings using smart meter data

Eline Himpe (UGent) and Arnold Janssens (UGent)
Author
Organization
Abstract
The increasing application of smart and digital energy meters leads to an increasing availability of frequent -e.g. hourly- and long-term measurements of the actual energy use in occupied buildings. In the resulting energy use time series, the diurnal fluctuations in energy use are recognised and similarities between diurnal profiles for various days are observed. These recurring profiles are called energy use time patterns and they are a result of various phenomena, such as patterns in the building use, occupational schedules, settings of the system control, short-term weather dynamics etc. These energy use time patterns can provide a better understanding of the energy use, which is useful in many fields including energy feedback, fault detection and energy auditing. In order to identify and characterise energy use time patterns for large data-sets, an automated approach is needed. This paper proposes a methodology for automated mathematical recognition of energy use time patterns based on cluster analysis. Secondly, a methodology to characterise the identified patterns in function of external variables is proposed, using classification analysis techniques. The methodologies allow an automated identification and characterization of energy use time patterns, allowing a better understanding of the variations and changes in building energy use and their relation to weather conditions and calendar aspects.
Keywords
smart meter, identification, energy use time patterns, residential buildings, clustering analysis, energy performance, building energy use, normalisation

Downloads

  • e3sconf clima2019 05011.pdf
    • full text (Published version)
    • |
    • open access
    • |
    • PDF
    • |
    • 2.02 MB

Citation

Please use this url to cite or link to this publication:

MLA
Himpe, Eline, and Arnold Janssens. “Identification of Energy Use Time Patterns of Occupied Dwellings Using Smart Meter Data.” E3S Web of Conferences, vol. 111, EDP Sciences, 2019.
APA
Himpe, E., & Janssens, A. (2019). Identification of energy use time patterns of occupied dwellings using smart meter data. In E3S Web of Conferences (Vol. 111). France: EDP Sciences.
Chicago author-date
Himpe, Eline, and Arnold Janssens. 2019. “Identification of Energy Use Time Patterns of Occupied Dwellings Using Smart Meter Data.” In E3S Web of Conferences. Vol. 111. France: EDP Sciences.
Chicago author-date (all authors)
Himpe, Eline, and Arnold Janssens. 2019. “Identification of Energy Use Time Patterns of Occupied Dwellings Using Smart Meter Data.” In E3S Web of Conferences. Vol. 111. France: EDP Sciences.
Vancouver
1.
Himpe E, Janssens A. Identification of energy use time patterns of occupied dwellings using smart meter data. In: E3S Web of Conferences. France: EDP Sciences; 2019.
IEEE
[1]
E. Himpe and A. Janssens, “Identification of energy use time patterns of occupied dwellings using smart meter data,” in E3S Web of Conferences, Bucharest, Romania, 2019, vol. 111.
@inproceedings{8635402,
  abstract     = {The increasing application of smart and digital energy meters leads to an increasing availability of frequent -e.g. hourly- and long-term measurements of the actual energy use in occupied buildings. In the resulting energy use time series, the diurnal fluctuations in energy use are recognised and similarities between diurnal profiles for various days are observed. These recurring profiles are called energy use time patterns and they are a result of various phenomena, such as patterns in the building use, occupational schedules, settings of the system control, short-term weather dynamics etc. These energy use time patterns can provide a better understanding of the energy use, which is useful in many fields including energy feedback, fault detection and energy auditing. In order to identify and characterise energy use time patterns for large data-sets, an automated approach is needed. This paper proposes a methodology for automated mathematical recognition of energy use time patterns based on cluster analysis. Secondly, a methodology to characterise the identified patterns in function of external variables is proposed, using classification analysis techniques. The methodologies allow an automated identification and characterization of energy use time patterns, allowing a better understanding of the variations and changes in building energy use and their relation to weather conditions and calendar aspects.},
  articleno    = {05011},
  author       = {Himpe, Eline and Janssens, Arnold},
  booktitle    = {E3S Web of Conferences},
  issn         = {2267-1242},
  keywords     = {smart meter,identification,energy use time patterns,residential buildings,clustering analysis,energy performance,building energy use,normalisation},
  language     = {eng},
  location     = {Bucharest, Romania},
  pages        = {7},
  publisher    = {EDP Sciences},
  title        = {Identification of energy use time patterns of occupied dwellings using smart meter data},
  url          = {http://dx.doi.org/10.1051/e3sconf/201911105011},
  volume       = {111},
  year         = {2019},
}

Altmetric
View in Altmetric