A generic optimisation-based approach for improving non-intrusive load monitoring

Kanghang He, Dusan Jakovetic, Bochao Zhao, Vladimir Stankovic, Lina Stankovic, Samuel Cheng

Research output: Contribution to journalArticle

Abstract

The large-scale deployment of smart metering worldwide has ignited renewed interest in electrical load disaggregation, or non-intrusive load monitoring (NILM). Most NILM algorithms disaggregate one appliance at a time, remove the estimated appliance contribution from the total load, and then move on to disaggregate the next appliance. On one hand, this is efficient since multi-class classification is avoided and analytical models for each appliance can be developed independently of other appliances with the benefit of being transferred to unseen houses that have different sets of appliances. On the other hand, however, these methods can significantly under- or over- estimate the total consumption since they do not minimise the difference between the measured aggregate load and the sum of estimated individual loads. Motivated by minimising the latter difference without losing the benefits of existing NILM algorithms, we propose novel post-processing approaches for improving the accuracy of existing NILM. This is posed as an optimisation problem to refine the final NILM result using regularisation, based on the level of confidence in the original NILM output. First, we propose a heuristic method to solve this (combinatorial) boolean quadratic problem through relaxing zero-one constraint sets to compact zero-one intervals. Convex-based solutions, including norm-1, norm-2 and semi-definite programming-based relaxation, are proposed trading off accuracy and complexity. We demonstrate good performance of the proposed post-processing methods, applicable to any event-based NILM, compared with 4 state-of-the-art benchmarks, using public REFIT and REDD electrical load datasets.
LanguageEnglish
Pages6472-6480
Number of pages9
JournalIEEE Transactions on Smart Grid
Volume10
Issue number6
Early online date19 Mar 2019
DOIs
Publication statusPublished - 1 Nov 2019

Fingerprint

Monitoring
Heuristic methods
Processing
Analytical models

Keywords

  • energy disaggregation
  • smart metering
  • NILM

Cite this

@article{a678d49fff094badb8b7d2c3accfaccb,
title = "A generic optimisation-based approach for improving non-intrusive load monitoring",
abstract = "The large-scale deployment of smart metering worldwide has ignited renewed interest in electrical load disaggregation, or non-intrusive load monitoring (NILM). Most NILM algorithms disaggregate one appliance at a time, remove the estimated appliance contribution from the total load, and then move on to disaggregate the next appliance. On one hand, this is efficient since multi-class classification is avoided and analytical models for each appliance can be developed independently of other appliances with the benefit of being transferred to unseen houses that have different sets of appliances. On the other hand, however, these methods can significantly under- or over- estimate the total consumption since they do not minimise the difference between the measured aggregate load and the sum of estimated individual loads. Motivated by minimising the latter difference without losing the benefits of existing NILM algorithms, we propose novel post-processing approaches for improving the accuracy of existing NILM. This is posed as an optimisation problem to refine the final NILM result using regularisation, based on the level of confidence in the original NILM output. First, we propose a heuristic method to solve this (combinatorial) boolean quadratic problem through relaxing zero-one constraint sets to compact zero-one intervals. Convex-based solutions, including norm-1, norm-2 and semi-definite programming-based relaxation, are proposed trading off accuracy and complexity. We demonstrate good performance of the proposed post-processing methods, applicable to any event-based NILM, compared with 4 state-of-the-art benchmarks, using public REFIT and REDD electrical load datasets.",
keywords = "energy disaggregation, smart metering, NILM",
author = "Kanghang He and Dusan Jakovetic and Bochao Zhao and Vladimir Stankovic and Lina Stankovic and Samuel Cheng",
note = "{\circledC} 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, 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 component of this work in other works.",
year = "2019",
month = "11",
day = "1",
doi = "10.1109/TSG.2019.2906012",
language = "English",
volume = "10",
pages = "6472--6480",
journal = "IEEE Transactions on Smart Grid",
issn = "1949-3053",
number = "6",

}

A generic optimisation-based approach for improving non-intrusive load monitoring. / He, Kanghang; Jakovetic, Dusan; Zhao, Bochao; Stankovic, Vladimir; Stankovic, Lina; Cheng, Samuel.

In: IEEE Transactions on Smart Grid, Vol. 10, No. 6, 01.11.2019, p. 6472-6480.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A generic optimisation-based approach for improving non-intrusive load monitoring

AU - He, Kanghang

AU - Jakovetic, Dusan

AU - Zhao, Bochao

AU - Stankovic, Vladimir

AU - Stankovic, Lina

AU - Cheng, Samuel

N1 - © 2019 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, 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 component of this work in other works.

PY - 2019/11/1

Y1 - 2019/11/1

N2 - The large-scale deployment of smart metering worldwide has ignited renewed interest in electrical load disaggregation, or non-intrusive load monitoring (NILM). Most NILM algorithms disaggregate one appliance at a time, remove the estimated appliance contribution from the total load, and then move on to disaggregate the next appliance. On one hand, this is efficient since multi-class classification is avoided and analytical models for each appliance can be developed independently of other appliances with the benefit of being transferred to unseen houses that have different sets of appliances. On the other hand, however, these methods can significantly under- or over- estimate the total consumption since they do not minimise the difference between the measured aggregate load and the sum of estimated individual loads. Motivated by minimising the latter difference without losing the benefits of existing NILM algorithms, we propose novel post-processing approaches for improving the accuracy of existing NILM. This is posed as an optimisation problem to refine the final NILM result using regularisation, based on the level of confidence in the original NILM output. First, we propose a heuristic method to solve this (combinatorial) boolean quadratic problem through relaxing zero-one constraint sets to compact zero-one intervals. Convex-based solutions, including norm-1, norm-2 and semi-definite programming-based relaxation, are proposed trading off accuracy and complexity. We demonstrate good performance of the proposed post-processing methods, applicable to any event-based NILM, compared with 4 state-of-the-art benchmarks, using public REFIT and REDD electrical load datasets.

AB - The large-scale deployment of smart metering worldwide has ignited renewed interest in electrical load disaggregation, or non-intrusive load monitoring (NILM). Most NILM algorithms disaggregate one appliance at a time, remove the estimated appliance contribution from the total load, and then move on to disaggregate the next appliance. On one hand, this is efficient since multi-class classification is avoided and analytical models for each appliance can be developed independently of other appliances with the benefit of being transferred to unseen houses that have different sets of appliances. On the other hand, however, these methods can significantly under- or over- estimate the total consumption since they do not minimise the difference between the measured aggregate load and the sum of estimated individual loads. Motivated by minimising the latter difference without losing the benefits of existing NILM algorithms, we propose novel post-processing approaches for improving the accuracy of existing NILM. This is posed as an optimisation problem to refine the final NILM result using regularisation, based on the level of confidence in the original NILM output. First, we propose a heuristic method to solve this (combinatorial) boolean quadratic problem through relaxing zero-one constraint sets to compact zero-one intervals. Convex-based solutions, including norm-1, norm-2 and semi-definite programming-based relaxation, are proposed trading off accuracy and complexity. We demonstrate good performance of the proposed post-processing methods, applicable to any event-based NILM, compared with 4 state-of-the-art benchmarks, using public REFIT and REDD electrical load datasets.

KW - energy disaggregation

KW - smart metering

KW - NILM

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

U2 - 10.1109/TSG.2019.2906012

DO - 10.1109/TSG.2019.2906012

M3 - Article

VL - 10

SP - 6472

EP - 6480

JO - IEEE Transactions on Smart Grid

T2 - IEEE Transactions on Smart Grid

JF - IEEE Transactions on Smart Grid

SN - 1949-3053

IS - 6

ER -