Developing a Hybrid Model of Prediction and Classification Algorithms for Building Energy Consumption

Saeed BANIHASHEMI, Grace Ding, Jack Wang

Research output: A Conference proceeding or a Chapter in BookConference contribution

13 Citations (Scopus)
2 Downloads (Pure)

Abstract

Artificial intelligence algorithms have been applied separately or integrally for prediction, classification or optimization of buildings energy consumption. However, there is a salient gap in the literature on the investigation of hybrid objective function development for energy optimization problems including qualitative and quantitative datasets in their constructs. To tackle with this challenge, this paper presents a hybrid objective function of machine learning algorithms in optimizing energy consumption of residential buildings through considering both continuous and discrete parameters of energy simultaneously. To do this, a comprehensive dataset including significant parameters of building envelop, building design layout and HVAC was established, Artificial Neural Network as a prediction and Decision Tree as a classification algorithm were employed via cross-training ensemble equation to create the hybrid function and the model was finally validated via the weighted average of the error decomposed for the performance. The developed model could effectively enhance the accuracy of the objective functions used in the building energy prediction and optimization problems. Furthermore, the results of this novel approach resolved the inclusion issue of both continuous and discrete parameters of energy in a unified objective function without threatening the integrity and consistency of the building energy datasets.

Original languageEnglish
Title of host publicationEnergy Procedia
EditorsFiroz Alam
Place of PublicationNetherland
PublisherElsevier
Pages371-376
Number of pages6
Volume110
DOIs
Publication statusPublished - 2017
Externally publishedYes
EventInternational Conference on Energy and Power (ICEP) - Melbourne, Australia
Duration: 14 Dec 2016 → …

Publication series

NameEnergy Procedia
PublisherElsevier BV
ISSN (Print)1876-6102

Conference

ConferenceInternational Conference on Energy and Power (ICEP)
CountryAustralia
CityMelbourne
Period14/12/16 → …

Fingerprint

Energy utilization
Decision trees
Learning algorithms
Artificial intelligence
Learning systems
Neural networks

Cite this

BANIHASHEMI, S., Ding, G., & Wang, J. (2017). Developing a Hybrid Model of Prediction and Classification Algorithms for Building Energy Consumption. In F. Alam (Ed.), Energy Procedia (Vol. 110, pp. 371-376). (Energy Procedia). Netherland: Elsevier. https://doi.org/10.1016/j.egypro.2017.03.155
BANIHASHEMI, Saeed ; Ding, Grace ; Wang, Jack. / Developing a Hybrid Model of Prediction and Classification Algorithms for Building Energy Consumption. Energy Procedia. editor / Firoz Alam. Vol. 110 Netherland : Elsevier, 2017. pp. 371-376 (Energy Procedia).
@inproceedings{c2a1bb9db07a4191b8de18aa811ff570,
title = "Developing a Hybrid Model of Prediction and Classification Algorithms for Building Energy Consumption",
abstract = "Artificial intelligence algorithms have been applied separately or integrally for prediction, classification or optimization of buildings energy consumption. However, there is a salient gap in the literature on the investigation of hybrid objective function development for energy optimization problems including qualitative and quantitative datasets in their constructs. To tackle with this challenge, this paper presents a hybrid objective function of machine learning algorithms in optimizing energy consumption of residential buildings through considering both continuous and discrete parameters of energy simultaneously. To do this, a comprehensive dataset including significant parameters of building envelop, building design layout and HVAC was established, Artificial Neural Network as a prediction and Decision Tree as a classification algorithm were employed via cross-training ensemble equation to create the hybrid function and the model was finally validated via the weighted average of the error decomposed for the performance. The developed model could effectively enhance the accuracy of the objective functions used in the building energy prediction and optimization problems. Furthermore, the results of this novel approach resolved the inclusion issue of both continuous and discrete parameters of energy in a unified objective function without threatening the integrity and consistency of the building energy datasets.",
keywords = "Energy optimisation, machine learning, prediction, classification, residential buildings",
author = "Saeed BANIHASHEMI and Grace Ding and Jack Wang",
year = "2017",
doi = "10.1016/j.egypro.2017.03.155",
language = "English",
volume = "110",
series = "Energy Procedia",
publisher = "Elsevier",
pages = "371--376",
editor = "Firoz Alam",
booktitle = "Energy Procedia",
address = "Netherlands",

}

BANIHASHEMI, S, Ding, G & Wang, J 2017, Developing a Hybrid Model of Prediction and Classification Algorithms for Building Energy Consumption. in F Alam (ed.), Energy Procedia. vol. 110, Energy Procedia, Elsevier, Netherland, pp. 371-376, International Conference on Energy and Power (ICEP), Melbourne, Australia, 14/12/16. https://doi.org/10.1016/j.egypro.2017.03.155

Developing a Hybrid Model of Prediction and Classification Algorithms for Building Energy Consumption. / BANIHASHEMI, Saeed; Ding, Grace; Wang, Jack.

Energy Procedia. ed. / Firoz Alam. Vol. 110 Netherland : Elsevier, 2017. p. 371-376 (Energy Procedia).

Research output: A Conference proceeding or a Chapter in BookConference contribution

TY - GEN

T1 - Developing a Hybrid Model of Prediction and Classification Algorithms for Building Energy Consumption

AU - BANIHASHEMI, Saeed

AU - Ding, Grace

AU - Wang, Jack

PY - 2017

Y1 - 2017

N2 - Artificial intelligence algorithms have been applied separately or integrally for prediction, classification or optimization of buildings energy consumption. However, there is a salient gap in the literature on the investigation of hybrid objective function development for energy optimization problems including qualitative and quantitative datasets in their constructs. To tackle with this challenge, this paper presents a hybrid objective function of machine learning algorithms in optimizing energy consumption of residential buildings through considering both continuous and discrete parameters of energy simultaneously. To do this, a comprehensive dataset including significant parameters of building envelop, building design layout and HVAC was established, Artificial Neural Network as a prediction and Decision Tree as a classification algorithm were employed via cross-training ensemble equation to create the hybrid function and the model was finally validated via the weighted average of the error decomposed for the performance. The developed model could effectively enhance the accuracy of the objective functions used in the building energy prediction and optimization problems. Furthermore, the results of this novel approach resolved the inclusion issue of both continuous and discrete parameters of energy in a unified objective function without threatening the integrity and consistency of the building energy datasets.

AB - Artificial intelligence algorithms have been applied separately or integrally for prediction, classification or optimization of buildings energy consumption. However, there is a salient gap in the literature on the investigation of hybrid objective function development for energy optimization problems including qualitative and quantitative datasets in their constructs. To tackle with this challenge, this paper presents a hybrid objective function of machine learning algorithms in optimizing energy consumption of residential buildings through considering both continuous and discrete parameters of energy simultaneously. To do this, a comprehensive dataset including significant parameters of building envelop, building design layout and HVAC was established, Artificial Neural Network as a prediction and Decision Tree as a classification algorithm were employed via cross-training ensemble equation to create the hybrid function and the model was finally validated via the weighted average of the error decomposed for the performance. The developed model could effectively enhance the accuracy of the objective functions used in the building energy prediction and optimization problems. Furthermore, the results of this novel approach resolved the inclusion issue of both continuous and discrete parameters of energy in a unified objective function without threatening the integrity and consistency of the building energy datasets.

KW - Energy optimisation

KW - machine learning

KW - prediction

KW - classification

KW - residential buildings

UR - http://www.scopus.com/inward/record.url?scp=85019588397&partnerID=8YFLogxK

UR - http://www.mendeley.com/research/developing-hybrid-model-prediction-classification-algorithms-building-energy-consumption

U2 - 10.1016/j.egypro.2017.03.155

DO - 10.1016/j.egypro.2017.03.155

M3 - Conference contribution

VL - 110

T3 - Energy Procedia

SP - 371

EP - 376

BT - Energy Procedia

A2 - Alam, Firoz

PB - Elsevier

CY - Netherland

ER -

BANIHASHEMI S, Ding G, Wang J. Developing a Hybrid Model of Prediction and Classification Algorithms for Building Energy Consumption. In Alam F, editor, Energy Procedia. Vol. 110. Netherland: Elsevier. 2017. p. 371-376. (Energy Procedia). https://doi.org/10.1016/j.egypro.2017.03.155