A platform for research: civil engineering, architecture and urbanism
A platform for research: civil engineering, architecture and urbanism
A Data-Driven Approach to Predict Building Energy Performance for Identifying Optimal Energy Retrofit Scenarios
https://orcid.org/http://orcid.org/0000-0001-8025-6513
https://orcid.org/http://orcid.org/0000-0003-3931-0017
https://orcid.org/http://orcid.org/0000-0003-2634-0014
https://orcid.org/http://orcid.org/0000-0003-1474-2366
It is well established that building energy retrofits can improve building energy efficiency and reduce carbon emissions from the existing building sector in Canada. Furthermore, an accurate prediction of building energy performance can assist stakeholders in evaluating numerous retrofit scenarios and identifying optimal ones. This study aims to develop a data-driven approach to predict the energy performance of different retrofit scenarios. In this regard, four machine learning models were employed to predict the electricity and natural gas consumption in a typical Canadian single-family residential building. The investigated machine learning models were linear regression, support vector regression, artificial neural network, and extreme gradient boosting. In addition, energy consumption was used to realize the environmental and economic performance of various retrofit scenarios. Finally, Pareto optimization was used to identify optimal retrofit scenarios for the residential building. The results indicated that the proposed approach could be an alternative to complex building energy simulation tools to predict building energy performance and aid energy retrofit decision-making.
A Data-Driven Approach to Predict Building Energy Performance for Identifying Optimal Energy Retrofit Scenarios
https://orcid.org/http://orcid.org/0000-0001-8025-6513
https://orcid.org/http://orcid.org/0000-0003-3931-0017
https://orcid.org/http://orcid.org/0000-0003-2634-0014
https://orcid.org/http://orcid.org/0000-0003-1474-2366
It is well established that building energy retrofits can improve building energy efficiency and reduce carbon emissions from the existing building sector in Canada. Furthermore, an accurate prediction of building energy performance can assist stakeholders in evaluating numerous retrofit scenarios and identifying optimal ones. This study aims to develop a data-driven approach to predict the energy performance of different retrofit scenarios. In this regard, four machine learning models were employed to predict the electricity and natural gas consumption in a typical Canadian single-family residential building. The investigated machine learning models were linear regression, support vector regression, artificial neural network, and extreme gradient boosting. In addition, energy consumption was used to realize the environmental and economic performance of various retrofit scenarios. Finally, Pareto optimization was used to identify optimal retrofit scenarios for the residential building. The results indicated that the proposed approach could be an alternative to complex building energy simulation tools to predict building energy performance and aid energy retrofit decision-making.
A Data-Driven Approach to Predict Building Energy Performance for Identifying Optimal Energy Retrofit Scenarios
https://orcid.org/http://orcid.org/0000-0001-8025-6513
https://orcid.org/http://orcid.org/0000-0003-3931-0017
https://orcid.org/http://orcid.org/0000-0003-2634-0014
https://orcid.org/http://orcid.org/0000-0003-1474-2366
It is well established that building energy retrofits can improve building energy efficiency and reduce carbon emissions from the existing building sector in Canada. Furthermore, an accurate prediction of building energy performance can assist stakeholders in evaluating numerous retrofit scenarios and identifying optimal ones. This study aims to develop a data-driven approach to predict the energy performance of different retrofit scenarios. In this regard, four machine learning models were employed to predict the electricity and natural gas consumption in a typical Canadian single-family residential building. The investigated machine learning models were linear regression, support vector regression, artificial neural network, and extreme gradient boosting. In addition, energy consumption was used to realize the environmental and economic performance of various retrofit scenarios. Finally, Pareto optimization was used to identify optimal retrofit scenarios for the residential building. The results indicated that the proposed approach could be an alternative to complex building energy simulation tools to predict building energy performance and aid energy retrofit decision-making.
A Data-Driven Approach to Predict Building Energy Performance for Identifying Optimal Energy Retrofit Scenarios
Lecture Notes in Civil Engineering
Desjardins, Serge (editor) / Poitras, Gérard J. (editor) / Zhang, Haonan (author) / Hewage, Kasun (author) / Hussain, Syed Asad (author) / Sadiq, Rehan (author)
Canadian Society of Civil Engineering Annual Conference ; 2023 ; Moncton, NB, Canada
2024-10-01
13 pages
Article/Chapter (Book)
Electronic Resource
English
| Elsevier | 2024
| Taylor & Francis Verlag | 2015