Evaluation of the influence of the bus-bar voltage profiles on demand forecasting by using neural networks
Autor(a) principal: | |
---|---|
Data de Publicação: | 2021 |
Outros Autores: | , , , , |
Tipo de documento: | Artigo |
Idioma: | por |
Título da fonte: | Research, Society and Development |
Texto Completo: | https://rsdjournal.org/index.php/rsd/article/view/20917 |
Resumo: | The very short-term load forecasting allows operation engineers an economic and safe dispatch of the power system while dynamically contributes to the prices in the energy market. Several methodologies such as regression analysis, time series, machine learning approaches, deep learning methods, and artificial intelligence have been used to forecast load. However, several external factors become the forecasting a more complex task than it initially appears to be. For this reason, neural networks have been presented as a computational intelligence technique capable of dealing with the load forecasting problem with great precision. In this context, this work aims to evaluate the impact of voltage profiles of the power system bus on the load forecasting. For this, it was proposed to study three database arrangements ((1) normalized load historical data; (2) normalized load historical data and voltage profile in load bars; and, (3) normalized load historical data, voltage profile in load bus and seasonality of these loads) to train nine neural networks of the MLP type with two layers. The proposed approach is evaluated based on data obtained from the state estimator of a network of a large company in the northern region of Brazil. The results show that, according to the MSE and MAPE values obtained, all the neural networks evaluated achieve a forecasting of the load as expected. However, the best performance was achieved with the arrangement that considered a database that records a normalized load historical data and voltage profile in load bus. |
id |
UNIFEI_6111cb1c1ebebcb520490ea8046f86d2 |
---|---|
oai_identifier_str |
oai:ojs.pkp.sfu.ca:article/20917 |
network_acronym_str |
UNIFEI |
network_name_str |
Research, Society and Development |
repository_id_str |
|
spelling |
Evaluation of the influence of the bus-bar voltage profiles on demand forecasting by using neural networksEvaluación de la influencia de los voltajes de barra en el pronóstico de la demanda mediante el uso de redes neuronalesAvaliação da influência das tensões de barra na previsão de cargas via redes neuraisPronóstico de demandaMuy cirto plazo redes neuronalesHistórico de cargaPerfiles de tensión.Load forecastingVery short-termNeural networksLoad historical dataVoltage profile.Previsão de cargaRedes neuraisTensão nas barrasHistórico de cargaCurtíssimo prazo.The very short-term load forecasting allows operation engineers an economic and safe dispatch of the power system while dynamically contributes to the prices in the energy market. Several methodologies such as regression analysis, time series, machine learning approaches, deep learning methods, and artificial intelligence have been used to forecast load. However, several external factors become the forecasting a more complex task than it initially appears to be. For this reason, neural networks have been presented as a computational intelligence technique capable of dealing with the load forecasting problem with great precision. In this context, this work aims to evaluate the impact of voltage profiles of the power system bus on the load forecasting. For this, it was proposed to study three database arrangements ((1) normalized load historical data; (2) normalized load historical data and voltage profile in load bars; and, (3) normalized load historical data, voltage profile in load bus and seasonality of these loads) to train nine neural networks of the MLP type with two layers. The proposed approach is evaluated based on data obtained from the state estimator of a network of a large company in the northern region of Brazil. The results show that, according to the MSE and MAPE values obtained, all the neural networks evaluated achieve a forecasting of the load as expected. However, the best performance was achieved with the arrangement that considered a database that records a normalized load historical data and voltage profile in load bus.El pronóstico de demanda de muy corto plazo permite a los ingenieros de operación un despacho económico y seguro del sistema eléctrico, además de ayudar en la composición dinámica de los precios en el mercado de energía. Varias metodologías como análisis de regresión, series temporales, enfoques de aprendizaje de máquina, métodos de aprendizaje profundo e inteligencia artificial han sido usadas para pronosticar la demanda. Sin embargo, varios factores externos tornan el pronóstico una tarea más compleja de lo que inicialmente aparenta ser. Es por ello, que las redes neuronales se han presentados como una técnica de inteligencia computacional capaz de lidiar con el problema de la previsión de carga con una grande precisión. En este contexto, este trabajo pretende evaluar el impacto de los perfiles de tensión de las barras del sistema eléctrico sobre el pronóstico de la demanda. Para ello, fue propuesto estudiar tres arreglos de bases de datos ((1) datos normalizados del histórico de carga; (2) datos normalizados del histórico de carga y tensión en las barras de carga; e, (3) datos normalizados del histórico de carga, tensión en las barras de carga y estacionalidad de estas) para entrenar nueve redes neuronales del tipo MLP con dos capas. El enfoque propuesto es evaluado con base en datos exportados por el estimador de estado de una red de una gran empresa de la región norte de Brasil. Los resultados indican que, según los valores de MSE y MAPE obtenidos, todas las redes neuronales evaluadas consiguen el pronóstico esperado. No obstante, el mejor desempeño fue alcanzado con el arreglo que consideró una base de datos a partir del histórico de carga normalizado y la tensión en las barras de carga.A previsão de carga de curtíssimo prazo permite aos engenheiros de operação um despacho econômico e seguro do sistema elétrico, além de ajudar na composição dinâmica de preços no mercado de energia. Diversas metodologias como análise de regressão, series temporais, abordagens de aprendizado de máquina, métodos de aprendizado profundo e inteligência artificial tem sido usadas para prever a carga. Mas, diversos fatores externos tornam a previsão uma tarefa mais complexa do que aparenta ser inicialmente. Assim, as redes neurais artificiais têm-se apresentado como uma técnica de inteligência computacional capaz de lidar com o problema da previsão de carga com grande precisão. Neste contexto, este trabalho visa avaliar o impacto do perfil de tensão das barras do sistema elétrico sobre a previsão da carga. Para isto, foi proposto estudar três arranjos de bases de dados ((1) dados normalizados do histórico de carga; (2) dados normalizados do histórico de carga e tensão nas barras de carga; e, (3) dados normalizados do histórico de carga, tensão nas barras de carga e sazonalidade destes) para treinar nove redes neurais do tipo MLP de duas camadas. A abordagem proposta é avaliada com base em dados exportados pelo estimador de estados da rede de uma grande empresa da região norte do Brasil. Os resultados indicam que, segundo os valores de MSE e MAPE obtidos, todas as redes neurais avaliadas atingem a previsão esperada. No entanto, o melhor desempenho foi alcançado com o arranjo que considerou uma base de dados a partir do histórico de carga normalizado e tensão nas barras de carga.Research, Society and Development2021-10-02info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfhttps://rsdjournal.org/index.php/rsd/article/view/2091710.33448/rsd-v10i12.20917Research, Society and Development; Vol. 10 No. 12; e600101220917Research, Society and Development; Vol. 10 Núm. 12; e600101220917Research, Society and Development; v. 10 n. 12; e6001012209172525-3409reponame:Research, Society and Developmentinstname:Universidade Federal de Itajubá (UNIFEI)instacron:UNIFEIporhttps://rsdjournal.org/index.php/rsd/article/view/20917/18585Copyright (c) 2021 Amanda Thais dos Reis Fernandes; Jean Lucas Tourinho Fonseca; Iuri Leno Pereira da Silva; Piedy Del Mar Agamez Arias; Rodrigo Andrade Ramos; Werbeston Douglas de Oliveirahttps://creativecommons.org/licenses/by/4.0info:eu-repo/semantics/openAccessFernandes, Amanda Thais dos Reis Fonseca, Jean Lucas Tourinho Silva, Iuri Leno Pereira da Agamez Arias, Piedy Del Mar Ramos, Rodrigo Andrade Oliveira, Werbeston Douglas de 2021-11-14T20:26:51Zoai:ojs.pkp.sfu.ca:article/20917Revistahttps://rsdjournal.org/index.php/rsd/indexPUBhttps://rsdjournal.org/index.php/rsd/oairsd.articles@gmail.com2525-34092525-3409opendoar:2024-01-17T09:40:25.889856Research, Society and Development - Universidade Federal de Itajubá (UNIFEI)false |
dc.title.none.fl_str_mv |
Evaluation of the influence of the bus-bar voltage profiles on demand forecasting by using neural networks Evaluación de la influencia de los voltajes de barra en el pronóstico de la demanda mediante el uso de redes neuronales Avaliação da influência das tensões de barra na previsão de cargas via redes neurais |
title |
Evaluation of the influence of the bus-bar voltage profiles on demand forecasting by using neural networks |
spellingShingle |
Evaluation of the influence of the bus-bar voltage profiles on demand forecasting by using neural networks Fernandes, Amanda Thais dos Reis Pronóstico de demanda Muy cirto plazo redes neuronales Histórico de carga Perfiles de tensión. Load forecasting Very short-term Neural networks Load historical data Voltage profile. Previsão de carga Redes neurais Tensão nas barras Histórico de carga Curtíssimo prazo. |
title_short |
Evaluation of the influence of the bus-bar voltage profiles on demand forecasting by using neural networks |
title_full |
Evaluation of the influence of the bus-bar voltage profiles on demand forecasting by using neural networks |
title_fullStr |
Evaluation of the influence of the bus-bar voltage profiles on demand forecasting by using neural networks |
title_full_unstemmed |
Evaluation of the influence of the bus-bar voltage profiles on demand forecasting by using neural networks |
title_sort |
Evaluation of the influence of the bus-bar voltage profiles on demand forecasting by using neural networks |
author |
Fernandes, Amanda Thais dos Reis |
author_facet |
Fernandes, Amanda Thais dos Reis Fonseca, Jean Lucas Tourinho Silva, Iuri Leno Pereira da Agamez Arias, Piedy Del Mar Ramos, Rodrigo Andrade Oliveira, Werbeston Douglas de |
author_role |
author |
author2 |
Fonseca, Jean Lucas Tourinho Silva, Iuri Leno Pereira da Agamez Arias, Piedy Del Mar Ramos, Rodrigo Andrade Oliveira, Werbeston Douglas de |
author2_role |
author author author author author |
dc.contributor.author.fl_str_mv |
Fernandes, Amanda Thais dos Reis Fonseca, Jean Lucas Tourinho Silva, Iuri Leno Pereira da Agamez Arias, Piedy Del Mar Ramos, Rodrigo Andrade Oliveira, Werbeston Douglas de |
dc.subject.por.fl_str_mv |
Pronóstico de demanda Muy cirto plazo redes neuronales Histórico de carga Perfiles de tensión. Load forecasting Very short-term Neural networks Load historical data Voltage profile. Previsão de carga Redes neurais Tensão nas barras Histórico de carga Curtíssimo prazo. |
topic |
Pronóstico de demanda Muy cirto plazo redes neuronales Histórico de carga Perfiles de tensión. Load forecasting Very short-term Neural networks Load historical data Voltage profile. Previsão de carga Redes neurais Tensão nas barras Histórico de carga Curtíssimo prazo. |
description |
The very short-term load forecasting allows operation engineers an economic and safe dispatch of the power system while dynamically contributes to the prices in the energy market. Several methodologies such as regression analysis, time series, machine learning approaches, deep learning methods, and artificial intelligence have been used to forecast load. However, several external factors become the forecasting a more complex task than it initially appears to be. For this reason, neural networks have been presented as a computational intelligence technique capable of dealing with the load forecasting problem with great precision. In this context, this work aims to evaluate the impact of voltage profiles of the power system bus on the load forecasting. For this, it was proposed to study three database arrangements ((1) normalized load historical data; (2) normalized load historical data and voltage profile in load bars; and, (3) normalized load historical data, voltage profile in load bus and seasonality of these loads) to train nine neural networks of the MLP type with two layers. The proposed approach is evaluated based on data obtained from the state estimator of a network of a large company in the northern region of Brazil. The results show that, according to the MSE and MAPE values obtained, all the neural networks evaluated achieve a forecasting of the load as expected. However, the best performance was achieved with the arrangement that considered a database that records a normalized load historical data and voltage profile in load bus. |
publishDate |
2021 |
dc.date.none.fl_str_mv |
2021-10-02 |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion |
format |
article |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
https://rsdjournal.org/index.php/rsd/article/view/20917 10.33448/rsd-v10i12.20917 |
url |
https://rsdjournal.org/index.php/rsd/article/view/20917 |
identifier_str_mv |
10.33448/rsd-v10i12.20917 |
dc.language.iso.fl_str_mv |
por |
language |
por |
dc.relation.none.fl_str_mv |
https://rsdjournal.org/index.php/rsd/article/view/20917/18585 |
dc.rights.driver.fl_str_mv |
https://creativecommons.org/licenses/by/4.0 info:eu-repo/semantics/openAccess |
rights_invalid_str_mv |
https://creativecommons.org/licenses/by/4.0 |
eu_rights_str_mv |
openAccess |
dc.format.none.fl_str_mv |
application/pdf |
dc.publisher.none.fl_str_mv |
Research, Society and Development |
publisher.none.fl_str_mv |
Research, Society and Development |
dc.source.none.fl_str_mv |
Research, Society and Development; Vol. 10 No. 12; e600101220917 Research, Society and Development; Vol. 10 Núm. 12; e600101220917 Research, Society and Development; v. 10 n. 12; e600101220917 2525-3409 reponame:Research, Society and Development instname:Universidade Federal de Itajubá (UNIFEI) instacron:UNIFEI |
instname_str |
Universidade Federal de Itajubá (UNIFEI) |
instacron_str |
UNIFEI |
institution |
UNIFEI |
reponame_str |
Research, Society and Development |
collection |
Research, Society and Development |
repository.name.fl_str_mv |
Research, Society and Development - Universidade Federal de Itajubá (UNIFEI) |
repository.mail.fl_str_mv |
rsd.articles@gmail.com |
_version_ |
1797052789578268672 |