Metrics
93 Record Views
Conference proceeding
Similarity measures in smart building electrical demand data
SoutheastCon 2015, Vol.2015, pp.1-4
SoutheastCon (Fort Lauderdale, FL, USA, 04/09/2015–04/12/2015)
04/2015
Web of Science ID: WOS:000371393800169
Abstract
With the increase in smart, LEED-certified buildings there comes an increase in the amount of time-series data generated by the sensor networks within these buildings. Extracting useful information from the sensor network data can pose a challenge. While diurnal and seasonal patterns of electrical demand are well known from traditional metering systems, smart building sensor networks can provide insight into abnormalities or previously unknown patterns in electrical demand. In this paper, we demonstrate how to mine the data for these unknowns through the analysis of the frequency components of the time-series electrical demand data. The data for this study was collected from an LEED-certified building over twelve consecutive months with separate data feeds for the electrical demand from the heating, A/C, ventilation, lighting and miscellaneous systems. We employed Fourier methods to transform the data from the time domain to the frequency domain and then used similarity measures to look for similarities and outliers among the differing systems.
Related links
Details
- Title
- Similarity measures in smart building electrical demand data
- Publication Details
- SoutheastCon 2015, Vol.2015, pp.1-4
- Resource Type
- Conference proceeding
- Conference
- SoutheastCon (Fort Lauderdale, FL, USA, 04/09/2015–04/12/2015)
- Publisher
- IEEE
- Copyright
- © 2015, IEEE
- Identifiers
- WOS:000371393800169; 99380178339806600
- Academic Unit
- Cybersecurity and Information Technology; Hal Marcus College of Science and Engineering
- Language
- English