skip to main content
OSTI.GOV title logo U.S. Department of Energy
Office of Scientific and Technical Information

Title: Comparison of Clustering Techniques for Residential Energy Behavior using Smart Meter Data

Abstract

Current practice in whole time series clustering of residential meter data focuses on aggregated or subsampled load data at the customer level, which ignores day-to-day differences within customers. This information is critical to determine each customer’s suitability to various demand side management strategies that support intelligent power grids and smart energy management. Clustering daily load shapes provides fine-grained information on customer attributes and sources of variation for subsequent models and customer segmentation. In this paper, we apply 11 clustering methods to daily residential meter data. We evaluate their parameter settings and suitability based on 6 generic performance metrics and post-checking of resulting clusters. Finally, we recommend suitable techniques and parameters based on the goal of discovering diverse daily load patterns among residential customers. To the authors’ knowledge, this paper is the first robust comparative review of clustering techniques applied to daily residential load shape time series in the power systems’ literature.

Authors:
; ; ; ; ; ;
Publication Date:
Research Org.:
Lawrence Berkeley National Lab. (LBNL), Berkeley, CA (United States)
Sponsoring Org.:
USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR) (SC-21)
OSTI Identifier:
1398467
DOE Contract Number:
AC02-05CH11231
Resource Type:
Conference
Resource Relation:
Conference: AAAI-17 Workshop on artificial intelligence for smart grids and smart buildings, San Francisco, CA (United States), 4-5 Feb 2017
Country of Publication:
United States
Language:
English
Subject:
24 POWER TRANSMISSION AND DISTRIBUTION

Citation Formats

Jin, Ling, Lee, Doris, Sim, Alex, Borgeson, Sam, Wu, Kesheng, Spurlock, C. Anna, and Todd, Annika. Comparison of Clustering Techniques for Residential Energy Behavior using Smart Meter Data. United States: N. p., 2017. Web.
Jin, Ling, Lee, Doris, Sim, Alex, Borgeson, Sam, Wu, Kesheng, Spurlock, C. Anna, & Todd, Annika. Comparison of Clustering Techniques for Residential Energy Behavior using Smart Meter Data. United States.
Jin, Ling, Lee, Doris, Sim, Alex, Borgeson, Sam, Wu, Kesheng, Spurlock, C. Anna, and Todd, Annika. Tue . "Comparison of Clustering Techniques for Residential Energy Behavior using Smart Meter Data". United States. doi:. https://www.osti.gov/servlets/purl/1398467.
@article{osti_1398467,
title = {Comparison of Clustering Techniques for Residential Energy Behavior using Smart Meter Data},
author = {Jin, Ling and Lee, Doris and Sim, Alex and Borgeson, Sam and Wu, Kesheng and Spurlock, C. Anna and Todd, Annika},
abstractNote = {Current practice in whole time series clustering of residential meter data focuses on aggregated or subsampled load data at the customer level, which ignores day-to-day differences within customers. This information is critical to determine each customer’s suitability to various demand side management strategies that support intelligent power grids and smart energy management. Clustering daily load shapes provides fine-grained information on customer attributes and sources of variation for subsequent models and customer segmentation. In this paper, we apply 11 clustering methods to daily residential meter data. We evaluate their parameter settings and suitability based on 6 generic performance metrics and post-checking of resulting clusters. Finally, we recommend suitable techniques and parameters based on the goal of discovering diverse daily load patterns among residential customers. To the authors’ knowledge, this paper is the first robust comparative review of clustering techniques applied to daily residential load shape time series in the power systems’ literature.},
doi = {},
journal = {},
number = ,
volume = ,
place = {United States},
year = {Tue Mar 21 00:00:00 EDT 2017},
month = {Tue Mar 21 00:00:00 EDT 2017}
}

Conference:
Other availability
Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:
  • We propose a method to examine structural and behavioral determinants of residential electricity consumption, by developing separate models for daily maximum (peak) and minimum (idle) consumption. We apply our method on a data set of 1628 households' electricity consumption. The results show that weather, location and floor area are among the most important determinants of residential electricity consumption. In addition to these variables, number of refrigerators and entertainment devices (e.g., VCRs) are among the most important determinants of daily minimum consumption, while number of occupants and high-consumption appliances such as electric water heaters are the most significant determinants of dailymore » maximum consumption. Installing double-pane windows and energy-efficient lights helped to reduce consumption, as did the energy-conscious use of electric heater. Acknowledging climate change as a motivation to save energy showed correlation with lower electricity consumption. Households with individuals over 55 or between 19 and 35 years old recorded lower electricity consumption, while pet owners showed higher consumption. Contrary to some previous studies, we observed no significant correlation between electricity consumption and income level, home ownership, or building age. Some otherwise energy-efficient features such as energy-efficient appliances, programmable thermostats, and insulation were correlated with slight increase in electricity consumption. (C) 2013 Elsevier Ltd. All rights reserved.« less
  • In the present work the measurement and analysis of pressure transients were used for leak detection and location in pipelines. Pressure transient profiles were detected in two pipelines, with 433 m and 20 m in length, by pressure transducers connected to PC computers equipped with A/D and D/A converters. Experiments were conducted for a wide range of liquid flow rates (Re from 3000 to 13000), leak positions and leak flow rates in the pipeline. The results show that leaks as small as 5 % of the nominal liquid flow rate are readily detected. The computed leak positions approached the actualmore » values with an error of 5 meters or less. 3 refs., 3 figs., 1 tab.« less
  • With the widespread deployment of electronic interval meters, commonly known as smart meters, came the promise of real-time data on electric energy consumption. Recognizing an opportunity to provide consumers access to their near real-time energy consumption data directly from their installed smart meter, we designed a mechanism for capturing those data for consumer use via an open smart energy gateway (OpenSEG). By design, OpenSEG provides a clearly defined boundary for equipment and data ownership. OpenSEG is an open-source data management platform to enable better data management of smart meter data. Effectively, it is an information architecture designed to work withmore » the ZigBee Smart Energy Profile 1.x (SEP 1.x). It was specifically designed to reduce cyber-security risks and provide secure information directly from smart meters to consumers in near real time, using display devices already owned by the consumers. OpenSEG stores 48 hours of recent consumption data in a circular cache using a format consistent with commonly available archived (not real-time) consumption data such as Green Button, which is based on the Energy Services Provider Interface (ESPI) data standard. It consists of a common XML format for energy usage information and a data exchange protocol to facilitate automated data transfer upon utility customer authorization. Included in the design is an application program interface by which users can acquire data from OpenSEG for further post processing. A sample data display application is included in the initial software product. The data display application demonstrates that OpenSEG can help electricity use data to be retrieved from a smart meter and ported to a wide variety of user-owned devices such as cell phones or a user-selected database. This system can be used for homes, multi-family buildings, or small commercial buildings in California.« less
  • Abstract not provided.