BiGRU-CNN neural network applied to short-term electric load forecasting
Autor(a) principal: | |
---|---|
Data de Publicação: | 2022 |
Outros Autores: | |
Tipo de documento: | Artigo |
Idioma: | eng |
Título da fonte: | Production |
Texto Completo: | http://old.scielo.br/scielo.php?script=sci_arttext&pid=S0103-65132022000100201 |
Resumo: | Abstract Paper aims This study analyzed the feasibility of the BiGRU-CNN artificial neural network as a forecasting tool for short-term electric load. This forecasting model can serve as a support tool related to decision-making by companies in the energy sector. Originality Despite a large amount of scientific research in this area, the literature still searches for more assertive forecasting models regarding short-term electric load. Thus, the BiGRU-CNN model, based on layers of BiGRU and CNN architecture networks was tested. This model was already proposed and used for other similar tasks, however, it has not been used on load forecasting. Research method The code was programmed in Python using the keras package. The forecasts of all networks were carried out 10 times until an acceptable statistical sample was reached so that future electric load values are as close as possible to reality. Main findings The best forecasting model was the proposed BiGRU-CNN network when compared to classical and some hybrid networks. Implications for theory and practice This methodology can be applied to short-term electric load forecasting problems. There is evidence that the combination of different layers of neural networks can provide more efficient forecasting results than classical networks with only one architecture. |
id |
ABEPRO-1_b9a16326929cae75e7ad3665b9e3a779 |
---|---|
oai_identifier_str |
oai:scielo:S0103-65132022000100201 |
network_acronym_str |
ABEPRO-1 |
network_name_str |
Production |
repository_id_str |
|
spelling |
BiGRU-CNN neural network applied to short-term electric load forecastingTime series forecastingRecurrent neural networksArtificial intelligenceMachine learningAbstract Paper aims This study analyzed the feasibility of the BiGRU-CNN artificial neural network as a forecasting tool for short-term electric load. This forecasting model can serve as a support tool related to decision-making by companies in the energy sector. Originality Despite a large amount of scientific research in this area, the literature still searches for more assertive forecasting models regarding short-term electric load. Thus, the BiGRU-CNN model, based on layers of BiGRU and CNN architecture networks was tested. This model was already proposed and used for other similar tasks, however, it has not been used on load forecasting. Research method The code was programmed in Python using the keras package. The forecasts of all networks were carried out 10 times until an acceptable statistical sample was reached so that future electric load values are as close as possible to reality. Main findings The best forecasting model was the proposed BiGRU-CNN network when compared to classical and some hybrid networks. Implications for theory and practice This methodology can be applied to short-term electric load forecasting problems. There is evidence that the combination of different layers of neural networks can provide more efficient forecasting results than classical networks with only one architecture.Associação Brasileira de Engenharia de Produção2022-01-01info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersiontext/htmlhttp://old.scielo.br/scielo.php?script=sci_arttext&pid=S0103-65132022000100201Production v.32 2022reponame:Productioninstname:Associação Brasileira de Engenharia de Produção (ABEPRO)instacron:ABEPRO10.1590/0103-6513.20210087info:eu-repo/semantics/openAccessSoares,Lucas DuarteFranco,Edgar Manuel Carreñoeng2021-12-03T00:00:00Zoai:scielo:S0103-65132022000100201Revistahttps://www.scielo.br/j/prod/https://old.scielo.br/oai/scielo-oai.php||production@editoracubo.com.br1980-54110103-6513opendoar:2021-12-03T00:00Production - Associação Brasileira de Engenharia de Produção (ABEPRO)false |
dc.title.none.fl_str_mv |
BiGRU-CNN neural network applied to short-term electric load forecasting |
title |
BiGRU-CNN neural network applied to short-term electric load forecasting |
spellingShingle |
BiGRU-CNN neural network applied to short-term electric load forecasting Soares,Lucas Duarte Time series forecasting Recurrent neural networks Artificial intelligence Machine learning |
title_short |
BiGRU-CNN neural network applied to short-term electric load forecasting |
title_full |
BiGRU-CNN neural network applied to short-term electric load forecasting |
title_fullStr |
BiGRU-CNN neural network applied to short-term electric load forecasting |
title_full_unstemmed |
BiGRU-CNN neural network applied to short-term electric load forecasting |
title_sort |
BiGRU-CNN neural network applied to short-term electric load forecasting |
author |
Soares,Lucas Duarte |
author_facet |
Soares,Lucas Duarte Franco,Edgar Manuel Carreño |
author_role |
author |
author2 |
Franco,Edgar Manuel Carreño |
author2_role |
author |
dc.contributor.author.fl_str_mv |
Soares,Lucas Duarte Franco,Edgar Manuel Carreño |
dc.subject.por.fl_str_mv |
Time series forecasting Recurrent neural networks Artificial intelligence Machine learning |
topic |
Time series forecasting Recurrent neural networks Artificial intelligence Machine learning |
description |
Abstract Paper aims This study analyzed the feasibility of the BiGRU-CNN artificial neural network as a forecasting tool for short-term electric load. This forecasting model can serve as a support tool related to decision-making by companies in the energy sector. Originality Despite a large amount of scientific research in this area, the literature still searches for more assertive forecasting models regarding short-term electric load. Thus, the BiGRU-CNN model, based on layers of BiGRU and CNN architecture networks was tested. This model was already proposed and used for other similar tasks, however, it has not been used on load forecasting. Research method The code was programmed in Python using the keras package. The forecasts of all networks were carried out 10 times until an acceptable statistical sample was reached so that future electric load values are as close as possible to reality. Main findings The best forecasting model was the proposed BiGRU-CNN network when compared to classical and some hybrid networks. Implications for theory and practice This methodology can be applied to short-term electric load forecasting problems. There is evidence that the combination of different layers of neural networks can provide more efficient forecasting results than classical networks with only one architecture. |
publishDate |
2022 |
dc.date.none.fl_str_mv |
2022-01-01 |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://old.scielo.br/scielo.php?script=sci_arttext&pid=S0103-65132022000100201 |
url |
http://old.scielo.br/scielo.php?script=sci_arttext&pid=S0103-65132022000100201 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.relation.none.fl_str_mv |
10.1590/0103-6513.20210087 |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
text/html |
dc.publisher.none.fl_str_mv |
Associação Brasileira de Engenharia de Produção |
publisher.none.fl_str_mv |
Associação Brasileira de Engenharia de Produção |
dc.source.none.fl_str_mv |
Production v.32 2022 reponame:Production instname:Associação Brasileira de Engenharia de Produção (ABEPRO) instacron:ABEPRO |
instname_str |
Associação Brasileira de Engenharia de Produção (ABEPRO) |
instacron_str |
ABEPRO |
institution |
ABEPRO |
reponame_str |
Production |
collection |
Production |
repository.name.fl_str_mv |
Production - Associação Brasileira de Engenharia de Produção (ABEPRO) |
repository.mail.fl_str_mv |
||production@editoracubo.com.br |
_version_ |
1754213154840117248 |