Data-driven approach to prediction of residential energy consumption at urban scales in London

Abdo Abdullah Ahmed Gassar, Geun Young Yun, Sumin Kim

Research output: Contribution to journalArticle

Abstract

Development of energy prediction models plays an integral part in management and enhancement of the energy efficiency of buildings, including carbon emission reduction. Simplified and data-driven models are often the preferred option when detailed information of simulation is not available and the fast responses are required. This study developed data-driven models for predicting electricity and gas consumption in London's residential buildings at the middle super output areas (MSOA) and lower super output areas (LSOA) with multilayer neural network (MNN), multiple regression (MLR), random forest (RF), and gradient boosting (GB) algorithms, and factors related to socio-demographic, economic, and building characteristics were used as predictors. The results revealed that building characteristics, household income, and the number of households were the most important predictors of electricity and gas consumption. We also found that MNN models have outperformed MLR, RF and GB models in electricity and gas consumption prediction at MSOA and LSOA levels, with R2 values over 0.99 for the electricity consumption model. In summary, this study shows that the MNN models can be a useful tool to assist the formation of energy efficiency policies in buildings at MSOA and LSOA levels.

Original languageEnglish
Article number115973
JournalEnergy
Volume187
DOIs
Publication statusPublished - 2019 Nov 15

Fingerprint

Energy utilization
Multilayer neural networks
Electricity
Energy efficiency
Gases
Economics
Carbon

All Science Journal Classification (ASJC) codes

  • Civil and Structural Engineering
  • Building and Construction
  • Pollution
  • Mechanical Engineering
  • Industrial and Manufacturing Engineering
  • Electrical and Electronic Engineering

Cite this

@article{d36fb4a3f8f64a14a875caab156b1b21,
title = "Data-driven approach to prediction of residential energy consumption at urban scales in London",
abstract = "Development of energy prediction models plays an integral part in management and enhancement of the energy efficiency of buildings, including carbon emission reduction. Simplified and data-driven models are often the preferred option when detailed information of simulation is not available and the fast responses are required. This study developed data-driven models for predicting electricity and gas consumption in London's residential buildings at the middle super output areas (MSOA) and lower super output areas (LSOA) with multilayer neural network (MNN), multiple regression (MLR), random forest (RF), and gradient boosting (GB) algorithms, and factors related to socio-demographic, economic, and building characteristics were used as predictors. The results revealed that building characteristics, household income, and the number of households were the most important predictors of electricity and gas consumption. We also found that MNN models have outperformed MLR, RF and GB models in electricity and gas consumption prediction at MSOA and LSOA levels, with R2 values over 0.99 for the electricity consumption model. In summary, this study shows that the MNN models can be a useful tool to assist the formation of energy efficiency policies in buildings at MSOA and LSOA levels.",
author = "{Ahmed Gassar}, {Abdo Abdullah} and Yun, {Geun Young} and Sumin Kim",
year = "2019",
month = "11",
day = "15",
doi = "10.1016/j.energy.2019.115973",
language = "English",
volume = "187",
journal = "Energy",
issn = "0360-5442",
publisher = "Elsevier Limited",

}

Data-driven approach to prediction of residential energy consumption at urban scales in London. / Ahmed Gassar, Abdo Abdullah; Yun, Geun Young; Kim, Sumin.

In: Energy, Vol. 187, 115973, 15.11.2019.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Data-driven approach to prediction of residential energy consumption at urban scales in London

AU - Ahmed Gassar, Abdo Abdullah

AU - Yun, Geun Young

AU - Kim, Sumin

PY - 2019/11/15

Y1 - 2019/11/15

N2 - Development of energy prediction models plays an integral part in management and enhancement of the energy efficiency of buildings, including carbon emission reduction. Simplified and data-driven models are often the preferred option when detailed information of simulation is not available and the fast responses are required. This study developed data-driven models for predicting electricity and gas consumption in London's residential buildings at the middle super output areas (MSOA) and lower super output areas (LSOA) with multilayer neural network (MNN), multiple regression (MLR), random forest (RF), and gradient boosting (GB) algorithms, and factors related to socio-demographic, economic, and building characteristics were used as predictors. The results revealed that building characteristics, household income, and the number of households were the most important predictors of electricity and gas consumption. We also found that MNN models have outperformed MLR, RF and GB models in electricity and gas consumption prediction at MSOA and LSOA levels, with R2 values over 0.99 for the electricity consumption model. In summary, this study shows that the MNN models can be a useful tool to assist the formation of energy efficiency policies in buildings at MSOA and LSOA levels.

AB - Development of energy prediction models plays an integral part in management and enhancement of the energy efficiency of buildings, including carbon emission reduction. Simplified and data-driven models are often the preferred option when detailed information of simulation is not available and the fast responses are required. This study developed data-driven models for predicting electricity and gas consumption in London's residential buildings at the middle super output areas (MSOA) and lower super output areas (LSOA) with multilayer neural network (MNN), multiple regression (MLR), random forest (RF), and gradient boosting (GB) algorithms, and factors related to socio-demographic, economic, and building characteristics were used as predictors. The results revealed that building characteristics, household income, and the number of households were the most important predictors of electricity and gas consumption. We also found that MNN models have outperformed MLR, RF and GB models in electricity and gas consumption prediction at MSOA and LSOA levels, with R2 values over 0.99 for the electricity consumption model. In summary, this study shows that the MNN models can be a useful tool to assist the formation of energy efficiency policies in buildings at MSOA and LSOA levels.

UR - http://www.scopus.com/inward/record.url?scp=85071090611&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071090611&partnerID=8YFLogxK

U2 - 10.1016/j.energy.2019.115973

DO - 10.1016/j.energy.2019.115973

M3 - Article

AN - SCOPUS:85071090611

VL - 187

JO - Energy

JF - Energy

SN - 0360-5442

M1 - 115973

ER -