Logo do repositório
 
Publicação

A Data-Driven Approach to Forecasting Heating and Cooling Energy Demand in an Office Building as an Alternative to Multi-Zone Dynamic Simulation

2021-02-03Artigo científico Acesso aberto
Ver registo simples
datacite.subject.fosEngenharia e Tecnologia
datacite.subject.fosCiências Naturais::Ciências Físicas
datacite.subject.fosEngenharia e Tecnologia::Engenharia Química
datacite.subject.fosCiências Naturais::Ciências da Computação e da Informação
datacite.subject.fosCiências Naturais::Outras Ciências Naturais
datacite.subject.sdg07:Energias Renováveis e Acessíveis
datacite.subject.sdg09:Indústria, Inovação e Infraestruturas
datacite.subject.sdg11:Cidades e Comunidades Sustentáveis
dc.contributor.authorGodinho, Xavier
dc.contributor.authorBernardo, Hermano
dc.contributor.authorSousa, João C. de
dc.contributor.authorOliveira, Filipe T.
dc.date.accessioned2026-04-08T13:48:16Z
dc.date.available2026-04-08T13:48:16Z
dc.date.issued2021-02-03
dc.description.abstractNowadays, as more data is now available from an increasing number of installed sensors, load forecasting applied to buildings is being increasingly explored. The amount and quality of resulting information can provide inputs for smarter decisions when managing and operating office buildings. In this article, the authors use two data-driven methods (artificial neural networks and support vector machines) to predict the heating and cooling energy demand in an office building located in Lisbon, Portugal. In the present case-study, these methods prove to be an accurate and appealing alternative to the use of accurate but time-consuming multi-zone dynamic simulation tools, which strongly depend on several parameters to be inserted and user expertise to calibrate the model. Artificial neural networks and support vector machines were developed and parametrized using historical data and different sets of exogenous variables to encounter the best performance combinations for both the heating and cooling periods of a year. In the case of support vector regression, a variation introduced simulated annealing to guide the search for different combinations of hyperparameters. After a feature selection stage for each individual method, the results for the different methods were compared, based on error metrics and distributions. The outputs of the study include the most suitable methodology for each season, and also the features (historical load records, but also exogenous features such as outdoor temperature, relative humidity or occupancy profile) that led to the most accurate models. Results clearly show there is a potential for faster, yet accurate machine-learning based forecasting methods to replace well-established, very accurate but time-consuming multi-zone dynamic simulation tools to forecast building energy consumption.eng
dc.description.sponsorshipThis research was partially supported by the European Regional Development Fund in the framework of COMPETE 2020 Programme through projects ESGRIDS (POCI-01-0145-FEDER-016434) and MAnAGER (POCI-01-0145-FEDER-028040), and the FCT—Portuguese Foundation for Science and Technology under project grant UIDB/00308/2020.
dc.identifier.citationGodinho, X.; Bernardo, H.; de Sousa, J.C.; Oliveira, F.T. A Data-Driven Approach to Forecasting Heating and Cooling Energy Demand in an Office Building as an Alternative to Multi-Zone Dynamic Simulation. Appl. Sci. 2021, 11, 1356. https://doi.org/10.3390/app11041356.
dc.identifier.doi10.3390/app11041356
dc.identifier.eissn2076-3417
dc.identifier.urihttp://hdl.handle.net/10400.8/16074
dc.language.isoeng
dc.peerreviewedyes
dc.publisherMDPI
dc.relationInstitute for Systems Engineering and Computers at Coimbra - INESC Coimbra
dc.relation.hasversionhttps://www.mdpi.com/2076-3417/11/4/1356
dc.relation.ispartofApplied Sciences
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.subjectload forecasting
dc.subjectbuilding energy demand
dc.subjectartificial neural networks
dc.subjectsupport vector machines
dc.subjectsimulated annealing
dc.subjectdata-driven methods
dc.subjectmulti-zone dynamic simulation
dc.titleA Data-Driven Approach to Forecasting Heating and Cooling Energy Demand in an Office Building as an Alternative to Multi-Zone Dynamic Simulationeng
dc.typejournal article
dspace.entity.typePublication
oaire.awardNumberUIDB/00308/2020
oaire.awardTitleInstitute for Systems Engineering and Computers at Coimbra - INESC Coimbra
oaire.awardURIinfo:eu-repo/grantAgreement/FCT/6817 - DCRRNI ID/UIDB%2F00308%2F2020/PT
oaire.citation.endPage23
oaire.citation.issue4
oaire.citation.startPage1
oaire.citation.titleApplied Sciences (Switzerland)
oaire.citation.volume11
oaire.fundingStream6817 - DCRRNI ID
oaire.versionhttp://purl.org/coar/version/c_970fb48d4fbd8a85
person.familyNameGodinho
person.familyNameBernardo
person.familyNameSousa
person.familyNameOliveira
person.givenNameXavier
person.givenNameHermano
person.givenNameJoão
person.givenNameFilipe Tadeu
person.identifier.ciencia-idC816-809B-4484
person.identifier.ciencia-id5A17-E4E1-C004
person.identifier.gsid55971568600
person.identifier.gsidmIvAC50AAAAJ
person.identifier.orcid0000-0001-8587-4812
person.identifier.orcid0000-0002-5290-6424
person.identifier.orcid0000-0002-7567-4910
person.identifier.orcid0000-0002-0410-4291
person.identifier.ridH-8020-2012
person.identifier.scopus-author-id57188647924
project.funder.identifierhttp://doi.org/10.13039/501100001871
project.funder.nameFundação para a Ciência e a Tecnologia
relation.isAuthorOfPublication8af70d06-e617-413b-8e51-2856604d0421
relation.isAuthorOfPublicationc58ec34e-5fbf-41cc-ae25-716ad0abb964
relation.isAuthorOfPublication7678f744-5e50-4458-8811-33e1fbc63013
relation.isAuthorOfPublicationf3c004bf-5759-4b8c-87c8-350bd598d035
relation.isAuthorOfPublication.latestForDiscovery8af70d06-e617-413b-8e51-2856604d0421
relation.isProjectOfPublication254d9223-2e3b-4754-bae9-c98986d80921
relation.isProjectOfPublication.latestForDiscovery254d9223-2e3b-4754-bae9-c98986d80921

Ficheiros

Principais
A mostrar 1 - 1 de 1
Miniatura
Nome:
A data-driven approach to forecasting heating and cooling energy demand in an office building as an alternative to multi-zone dynamic simulation.pdf
Tamanho:
10.78 MB
Formato:
Adobe Portable Document Format
Descrição:
Nowadays, as more data is now available from an increasing number of installed sensors, load forecasting applied to buildings is being increasingly explored. The amount and quality of resulting information can provide inputs for smarter decisions when managing and operating office buildings. In this article, the authors use two data-driven methods (artificial neural networks and support vector machines) to predict the heating and cooling energy demand in an office building located in Lisbon, Portugal. In the present case-study, these methods prove to be an accurate and appealing alternative to the use of accurate but time-consuming multi-zone dynamic simulation tools, which strongly depend on several parameters to be inserted and user expertise to calibrate the model. Artificial neural networks and support vector machines were developed and parametrized using historical data and different sets of exogenous variables to encounter the best performance combinations for both the heating and cooling periods of a year. In the case of support vector regression, a variation introduced simulated annealing to guide the search for different combinations of hyperparameters. After a feature selection stage for each individual method, the results for the different methods were compared, based on error metrics and distributions. The outputs of the study include the most suitable methodology for each season, and also the features (historical load records, but also exogenous features such as outdoor temperature, relative humidity or occupancy profile) that led to the most accurate models. Results clearly show there is a potential for faster, yet accurate machine-learning based forecasting methods to replace well-established, very accurate but time-consuming multi-zone dynamic simulation tools to forecast building energy consumption.
Ver/Abrir
Licença
A mostrar 1 - 1 de 1
Miniatura indisponível
Nome:
license.txt
Tamanho:
1.32 KB
Formato:
Item-specific license agreed upon to submission
Descrição:
Ver/Abrir

Coleções

INESCC-DL - Artigos em Revistas Internacionais
ESTG - Artigos em revistas internacionais