Non-intrusive load disaggregation using graph signal processing

Research output: Contribution to journalArticle

37 Citations (Scopus)

Abstract

With the large-scale roll-out of smart metering worldwide, there is a growing need to account for the individual contribution of appliances to the load demand. In this paper, we design a Graph signal processing (GSP)-based approach for non-intrusive appliance load monitoring (NILM), i.e., disaggregation of total energy consumption down to individual appliances used. Leveraging piecewise smoothness of the power load signal, two GSP-based NILM approaches are proposed. The first approach, based on total graph variation minimization, searches for a smooth graph signal under known label constraints. The second approach uses the total graph variation minimizer as a starting point for further refinement via simulated annealing. The proposed GSP-based NILM approach aims to address the large training overhead and associated complexity of conventional graph-based methods through a novel event-based graph approach. Simulation results using two datasets of real house measurements demonstrate the competitive performance of the GSP-based approaches with respect to traditionally used Hidden Markov Model-based and Decision Tree-based approaches.
LanguageEnglish
Pages1739-1747
Number of pages9
JournalIEEE Transactions on Smart Grid
Volume9
Issue number3
Early online date9 Aug 2016
DOIs
Publication statusPublished - 31 May 2018

Fingerprint

Signal processing
Monitoring
Hidden Markov models
Decision trees
Simulated annealing
Labels
Energy utilization

Keywords

  • energy disaggregation
  • graph signal processing
  • energy feedback
  • smart metering
  • appliances
  • load demand
  • energy consumption

Cite this

@article{98a549acfd9e4ad2a619c8c4b68fd320,
title = "Non-intrusive load disaggregation using graph signal processing",
abstract = "With the large-scale roll-out of smart metering worldwide, there is a growing need to account for the individual contribution of appliances to the load demand. In this paper, we design a Graph signal processing (GSP)-based approach for non-intrusive appliance load monitoring (NILM), i.e., disaggregation of total energy consumption down to individual appliances used. Leveraging piecewise smoothness of the power load signal, two GSP-based NILM approaches are proposed. The first approach, based on total graph variation minimization, searches for a smooth graph signal under known label constraints. The second approach uses the total graph variation minimizer as a starting point for further refinement via simulated annealing. The proposed GSP-based NILM approach aims to address the large training overhead and associated complexity of conventional graph-based methods through a novel event-based graph approach. Simulation results using two datasets of real house measurements demonstrate the competitive performance of the GSP-based approaches with respect to traditionally used Hidden Markov Model-based and Decision Tree-based approaches.",
keywords = "energy disaggregation, graph signal processing, energy feedback, smart metering, appliances, load demand, energy consumption",
author = "Kanghang He and Lina Stankovic and Jing Liao and Vladimir Stankovic",
note = "(c) 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.",
year = "2018",
month = "5",
day = "31",
doi = "10.1109/TSG.2016.2598872",
language = "English",
volume = "9",
pages = "1739--1747",
journal = "IEEE Transactions on Smart Grid",
issn = "1949-3053",
number = "3",

}

Non-intrusive load disaggregation using graph signal processing. / He, Kanghang; Stankovic, Lina; Liao, Jing; Stankovic, Vladimir.

In: IEEE Transactions on Smart Grid, Vol. 9, No. 3, 31.05.2018, p. 1739-1747.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Non-intrusive load disaggregation using graph signal processing

AU - He, Kanghang

AU - Stankovic, Lina

AU - Liao, Jing

AU - Stankovic, Vladimir

N1 - (c) 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works.

PY - 2018/5/31

Y1 - 2018/5/31

N2 - With the large-scale roll-out of smart metering worldwide, there is a growing need to account for the individual contribution of appliances to the load demand. In this paper, we design a Graph signal processing (GSP)-based approach for non-intrusive appliance load monitoring (NILM), i.e., disaggregation of total energy consumption down to individual appliances used. Leveraging piecewise smoothness of the power load signal, two GSP-based NILM approaches are proposed. The first approach, based on total graph variation minimization, searches for a smooth graph signal under known label constraints. The second approach uses the total graph variation minimizer as a starting point for further refinement via simulated annealing. The proposed GSP-based NILM approach aims to address the large training overhead and associated complexity of conventional graph-based methods through a novel event-based graph approach. Simulation results using two datasets of real house measurements demonstrate the competitive performance of the GSP-based approaches with respect to traditionally used Hidden Markov Model-based and Decision Tree-based approaches.

AB - With the large-scale roll-out of smart metering worldwide, there is a growing need to account for the individual contribution of appliances to the load demand. In this paper, we design a Graph signal processing (GSP)-based approach for non-intrusive appliance load monitoring (NILM), i.e., disaggregation of total energy consumption down to individual appliances used. Leveraging piecewise smoothness of the power load signal, two GSP-based NILM approaches are proposed. The first approach, based on total graph variation minimization, searches for a smooth graph signal under known label constraints. The second approach uses the total graph variation minimizer as a starting point for further refinement via simulated annealing. The proposed GSP-based NILM approach aims to address the large training overhead and associated complexity of conventional graph-based methods through a novel event-based graph approach. Simulation results using two datasets of real house measurements demonstrate the competitive performance of the GSP-based approaches with respect to traditionally used Hidden Markov Model-based and Decision Tree-based approaches.

KW - energy disaggregation

KW - graph signal processing

KW - energy feedback

KW - smart metering

KW - appliances

KW - load demand

KW - energy consumption

UR - http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?punumber=5165411

U2 - 10.1109/TSG.2016.2598872

DO - 10.1109/TSG.2016.2598872

M3 - Article

VL - 9

SP - 1739

EP - 1747

JO - IEEE Transactions on Smart Grid

T2 - IEEE Transactions on Smart Grid

JF - IEEE Transactions on Smart Grid

SN - 1949-3053

IS - 3

ER -