Low-complexity energy disaggregation using appliance load modelling

Hana Altrabalsi, Vladimir Stankovic, Jing Liao, Lina Stankovic

Research output: Contribution to journalArticle

Abstract

Large-scale smart metering deployments and energy saving targets across the world have ignited renewed interest in residential non-intrusive appliance load monitoring (NALM), that is, disaggregating total household's energy consumption down to individual appliances, using purely analytical tools. Despite increased research efforts, NALM techniques that can disaggregate power loads at low sampling rates are still not accurate and/or practical enough, requiring substantial customer input and long training periods. In this paper, we address these challenges via a practical low-complexity low-rate NALM, by proposing two approaches based on a combination of the following machine learning techniques: k-means clustering and Support Vector Machine, exploiting their strengths and addressing their individual weaknesses. The first proposed supervised approach is a low-complexity method that requires very short training period and is {fairly accurate even in the presence of} labelling errors. The second approach relies on a database of appliance signatures that we designed using publicly available datasets. The database compactly represents over 200 appliances using statistical modelling of measured active power. Experimental results on three datasets from US, Italy, Austria and UK, demonstrate the reliability and practicality.
LanguageEnglish
Pages884-905
Number of pages21
JournalAIMS Energy
Volume4
Issue number1
DOIs
Publication statusPublished - 5 Jan 2016

Fingerprint

Monitoring
Labeling
Support vector machines
Learning systems
Energy conservation
Energy utilization
Sampling

Keywords

  • energy disaggregation
  • appliance modelling
  • non-intrusive appliance load monitoring

Cite this

@article{18fa72c1593f4ae7bd14e58a92bd4502,
title = "Low-complexity energy disaggregation using appliance load modelling",
abstract = "Large-scale smart metering deployments and energy saving targets across the world have ignited renewed interest in residential non-intrusive appliance load monitoring (NALM), that is, disaggregating total household's energy consumption down to individual appliances, using purely analytical tools. Despite increased research efforts, NALM techniques that can disaggregate power loads at low sampling rates are still not accurate and/or practical enough, requiring substantial customer input and long training periods. In this paper, we address these challenges via a practical low-complexity low-rate NALM, by proposing two approaches based on a combination of the following machine learning techniques: k-means clustering and Support Vector Machine, exploiting their strengths and addressing their individual weaknesses. The first proposed supervised approach is a low-complexity method that requires very short training period and is {fairly accurate even in the presence of} labelling errors. The second approach relies on a database of appliance signatures that we designed using publicly available datasets. The database compactly represents over 200 appliances using statistical modelling of measured active power. Experimental results on three datasets from US, Italy, Austria and UK, demonstrate the reliability and practicality.",
keywords = "energy disaggregation, appliance modelling, non-intrusive appliance load monitoring",
author = "Hana Altrabalsi and Vladimir Stankovic and Jing Liao and Lina Stankovic",
year = "2016",
month = "1",
day = "5",
doi = "10.3934/energy.2016.1.1",
language = "English",
volume = "4",
pages = "884--905",
journal = "AIMS Energy",
issn = "2333-8326",
number = "1",

}

Low-complexity energy disaggregation using appliance load modelling. / Altrabalsi, Hana; Stankovic, Vladimir; Liao, Jing; Stankovic, Lina.

In: AIMS Energy, Vol. 4, No. 1, 05.01.2016, p. 884-905.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Low-complexity energy disaggregation using appliance load modelling

AU - Altrabalsi, Hana

AU - Stankovic, Vladimir

AU - Liao, Jing

AU - Stankovic, Lina

PY - 2016/1/5

Y1 - 2016/1/5

N2 - Large-scale smart metering deployments and energy saving targets across the world have ignited renewed interest in residential non-intrusive appliance load monitoring (NALM), that is, disaggregating total household's energy consumption down to individual appliances, using purely analytical tools. Despite increased research efforts, NALM techniques that can disaggregate power loads at low sampling rates are still not accurate and/or practical enough, requiring substantial customer input and long training periods. In this paper, we address these challenges via a practical low-complexity low-rate NALM, by proposing two approaches based on a combination of the following machine learning techniques: k-means clustering and Support Vector Machine, exploiting their strengths and addressing their individual weaknesses. The first proposed supervised approach is a low-complexity method that requires very short training period and is {fairly accurate even in the presence of} labelling errors. The second approach relies on a database of appliance signatures that we designed using publicly available datasets. The database compactly represents over 200 appliances using statistical modelling of measured active power. Experimental results on three datasets from US, Italy, Austria and UK, demonstrate the reliability and practicality.

AB - Large-scale smart metering deployments and energy saving targets across the world have ignited renewed interest in residential non-intrusive appliance load monitoring (NALM), that is, disaggregating total household's energy consumption down to individual appliances, using purely analytical tools. Despite increased research efforts, NALM techniques that can disaggregate power loads at low sampling rates are still not accurate and/or practical enough, requiring substantial customer input and long training periods. In this paper, we address these challenges via a practical low-complexity low-rate NALM, by proposing two approaches based on a combination of the following machine learning techniques: k-means clustering and Support Vector Machine, exploiting their strengths and addressing their individual weaknesses. The first proposed supervised approach is a low-complexity method that requires very short training period and is {fairly accurate even in the presence of} labelling errors. The second approach relies on a database of appliance signatures that we designed using publicly available datasets. The database compactly represents over 200 appliances using statistical modelling of measured active power. Experimental results on three datasets from US, Italy, Austria and UK, demonstrate the reliability and practicality.

KW - energy disaggregation

KW - appliance modelling

KW - non-intrusive appliance load monitoring

U2 - 10.3934/energy.2016.1.1

DO - 10.3934/energy.2016.1.1

M3 - Article

VL - 4

SP - 884

EP - 905

JO - AIMS Energy

T2 - AIMS Energy

JF - AIMS Energy

SN - 2333-8326

IS - 1

ER -