Energy Consumption Prediction Using Bands-Based Data Analytics

Kieran Greer; Y Bi

doi:10.1007/978-3-031-10989-8_50

Energy Consumption Prediction Using Bands-Based Data Analytics

Kieran Greer, Y Bi

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

77 Downloads (Pure)

Abstract

This paper describes a stochastic clustering method that is used for making predictions over energy data. The distinguishing feature of the method is discrete, localised optimisations based on similarity mea- surements, followed by a global aggregating layer, which can be compared with construction layers in deep neural networks. The developed model with the method is essentially a look-up table of the key energy bands that each appliance would use. Each band represents a level of consump- tion by the appliance. This table can replace disaggregation from more complicated methods, for instance constructed from probability theory. Experimental results show that the table can accurately disaggregate a single energy source to a set of appliances, because each appliance has quite a unique energy footprint. As part of predicting energy consump- tion, the model could possibly reduce costs by 50% and more than that if the proposed schedules are also included

Original language	English
Title of host publication	KNOWLEDGE SCIENCE, ENGINEERING AND MANAGEMENT, KSEM 2022, PT III
Editors	Gerard Memmi, Baijian Yang, Linghe Kong, Tianwei Zhang, Meikang Qiu
Pages	631-643
Number of pages	13
Volume	13370
ISBN (Electronic)	978-3-031-10989-8
DOIs	https://doi.org/10.1007/978-3-031-10989-8_50
Publication status	Published (in print/issue) - Aug 2022

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	13370 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Bibliographical note

Funding Information:
Acknowledgements. This work is supported by the project of “Novel Building Integration Designs for Increased Efficiencies in Advanced Climatically Tuneable Renewable Energy Systems (IDEAS)” (Grant ID: 815271), which is funded by the EU Horizon 2020 programme.

Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Keywords

Energy consumption prediction
Energy disaggregation
Stochastic clustering
Unsupervised machine learning

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1007/978-3-031-10989-8_50

Paper-Camera-Ready

Cite this

Greer, K., & Bi, Y. (2022). Energy Consumption Prediction Using Bands-Based Data Analytics. In G. Memmi, B. Yang, L. Kong, T. Zhang, & M. Qiu (Eds.), KNOWLEDGE SCIENCE, ENGINEERING AND MANAGEMENT, KSEM 2022, PT III (Vol. 13370, pp. 631-643). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13370 LNAI). https://doi.org/10.1007/978-3-031-10989-8_50

Greer, Kieran ; Bi, Y. / Energy Consumption Prediction Using Bands-Based Data Analytics. KNOWLEDGE SCIENCE, ENGINEERING AND MANAGEMENT, KSEM 2022, PT III. editor / Gerard Memmi ; Baijian Yang ; Linghe Kong ; Tianwei Zhang ; Meikang Qiu. Vol. 13370 2022. pp. 631-643 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{ffe1c7605d634f4eb7d06ee1056eee53,

title = "Energy Consumption Prediction Using Bands-Based Data Analytics",

abstract = "This paper describes a stochastic clustering method that is used for making predictions over energy data. The distinguishing feature of the method is discrete, localised optimisations based on similarity mea- surements, followed by a global aggregating layer, which can be compared with construction layers in deep neural networks. The developed model with the method is essentially a look-up table of the key energy bands that each appliance would use. Each band represents a level of consump- tion by the appliance. This table can replace disaggregation from more complicated methods, for instance constructed from probability theory. Experimental results show that the table can accurately disaggregate a single energy source to a set of appliances, because each appliance has quite a unique energy footprint. As part of predicting energy consump- tion, the model could possibly reduce costs by 50% and more than that if the proposed schedules are also included",

keywords = "Energy consumption prediction, Energy disaggregation, Stochastic clustering, Unsupervised machine learning",

author = "Kieran Greer and Y Bi",

note = "Funding Information: Acknowledgements. This work is supported by the project of “Novel Building Integration Designs for Increased Efficiencies in Advanced Climatically Tuneable Renewable Energy Systems (IDEAS)” (Grant ID: 815271), which is funded by the EU Horizon 2020 programme. Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.",

year = "2022",

month = aug,

doi = "10.1007/978-3-031-10989-8_50",

language = "English",

isbn = "9783031109881",

volume = "13370",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

pages = "631--643",

editor = "Gerard Memmi and Baijian Yang and Linghe Kong and Tianwei Zhang and Meikang Qiu",

booktitle = "KNOWLEDGE SCIENCE, ENGINEERING AND MANAGEMENT, KSEM 2022, PT III",

}

Greer, K & Bi, Y 2022, Energy Consumption Prediction Using Bands-Based Data Analytics. in G Memmi, B Yang, L Kong, T Zhang & M Qiu (eds), KNOWLEDGE SCIENCE, ENGINEERING AND MANAGEMENT, KSEM 2022, PT III. vol. 13370, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 13370 LNAI, pp. 631-643. https://doi.org/10.1007/978-3-031-10989-8_50

Energy Consumption Prediction Using Bands-Based Data Analytics. / Greer, Kieran; Bi, Y.
KNOWLEDGE SCIENCE, ENGINEERING AND MANAGEMENT, KSEM 2022, PT III. ed. / Gerard Memmi; Baijian Yang; Linghe Kong; Tianwei Zhang; Meikang Qiu. Vol. 13370 2022. p. 631-643 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 13370 LNAI).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Energy Consumption Prediction Using Bands-Based Data Analytics

AU - Greer, Kieran

AU - Bi, Y

N1 - Funding Information: Acknowledgements. This work is supported by the project of “Novel Building Integration Designs for Increased Efficiencies in Advanced Climatically Tuneable Renewable Energy Systems (IDEAS)” (Grant ID: 815271), which is funded by the EU Horizon 2020 programme. Publisher Copyright: © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

PY - 2022/8

Y1 - 2022/8

N2 - This paper describes a stochastic clustering method that is used for making predictions over energy data. The distinguishing feature of the method is discrete, localised optimisations based on similarity mea- surements, followed by a global aggregating layer, which can be compared with construction layers in deep neural networks. The developed model with the method is essentially a look-up table of the key energy bands that each appliance would use. Each band represents a level of consump- tion by the appliance. This table can replace disaggregation from more complicated methods, for instance constructed from probability theory. Experimental results show that the table can accurately disaggregate a single energy source to a set of appliances, because each appliance has quite a unique energy footprint. As part of predicting energy consump- tion, the model could possibly reduce costs by 50% and more than that if the proposed schedules are also included

AB - This paper describes a stochastic clustering method that is used for making predictions over energy data. The distinguishing feature of the method is discrete, localised optimisations based on similarity mea- surements, followed by a global aggregating layer, which can be compared with construction layers in deep neural networks. The developed model with the method is essentially a look-up table of the key energy bands that each appliance would use. Each band represents a level of consump- tion by the appliance. This table can replace disaggregation from more complicated methods, for instance constructed from probability theory. Experimental results show that the table can accurately disaggregate a single energy source to a set of appliances, because each appliance has quite a unique energy footprint. As part of predicting energy consump- tion, the model could possibly reduce costs by 50% and more than that if the proposed schedules are also included

KW - Energy consumption prediction

KW - Energy disaggregation

KW - Stochastic clustering

KW - Unsupervised machine learning

UR - http://www.scopus.com/inward/record.url?scp=85135006950&partnerID=8YFLogxK

U2 - 10.1007/978-3-031-10989-8_50

DO - 10.1007/978-3-031-10989-8_50

M3 - Conference contribution

SN - 9783031109881

VL - 13370

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 631

EP - 643

BT - KNOWLEDGE SCIENCE, ENGINEERING AND MANAGEMENT, KSEM 2022, PT III

A2 - Memmi, Gerard

A2 - Yang, Baijian

A2 - Kong, Linghe

A2 - Zhang, Tianwei

A2 - Qiu, Meikang

ER -

Greer K, Bi Y. Energy Consumption Prediction Using Bands-Based Data Analytics. In Memmi G, Yang B, Kong L, Zhang T, Qiu M, editors, KNOWLEDGE SCIENCE, ENGINEERING AND MANAGEMENT, KSEM 2022, PT III. Vol. 13370. 2022. p. 631-643. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-031-10989-8_50

Energy Consumption Prediction Using Bands-Based Data Analytics

Abstract

Publication series

Bibliographical note

Keywords

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this