A fast implementation of coalitional model predictive controllers based on machine learning: Application to solar power plants

This article proposes a real-time implementation of distributed model predictive controllers to maximize the thermal energy generated by parabolic trough collector fields. For this control strategy, we consider that each loop of the solar collector field is individually managed by a controller, whic...

Descripción completa

Detalles Bibliográficos
Autores: Masero Rubio, Eva, Ruiz-Moreno, Sara, Domínguez Frejo, José Ramón, Maestre Torreblanca, José María, Camacho, Eduardo F.
Tipo de recurso: artículo
Estado:Versión publicada
Fecha de publicación:2023
País:España
Institución:Universidad de Sevilla (US)
Repositorio:idUS. Depósito de Investigación de la Universidad de Sevilla
OAI Identifier:oai:idus.us.es:11441/141085
Acceso en línea:https://hdl.handle.net/11441/141085
https://doi.org/10.1016/j.engappai.2022.105666
Access Level:acceso abierto
Palabra clave:Neural networks
Artificial intelligence
Non-linear model predictive control
Coalitional control
Multi-agent systems
Solar thermal applications
id ES_7cfdca9d3c4f723ffbf90d242dc207c2
oai_identifier_str oai:idus.us.es:11441/141085
network_acronym_str ES
network_name_str España
repository_id_str
spelling A fast implementation of coalitional model predictive controllers based on machine learning: Application to solar power plantsMasero Rubio, EvaRuiz-Moreno, SaraDomínguez Frejo, José RamónMaestre Torreblanca, José MaríaCamacho, Eduardo F.Neural networksArtificial intelligenceNon-linear model predictive controlCoalitional controlMulti-agent systemsSolar thermal applicationsThis article proposes a real-time implementation of distributed model predictive controllers to maximize the thermal energy generated by parabolic trough collector fields. For this control strategy, we consider that each loop of the solar collector field is individually managed by a controller, which can form coalition with other controllers to attain its local goals while contributing to the overall objective. The formation of coalitions is based on a market-based mechanism in which the heat transfer fluid is traded. To relieve the computational burden online, we propose a learning-based approach that approximates optimization problems so that the controller can be applied in real time. Finally, simulations in a -loop solar collector field are used to assess the coalitional strategy based on neural networks in comparison with the coalitional model predictive control. The results show that the coalitional strategy based on neural networks provides a reduction in computing time of up to and a minimal reduction in performance compared to the coalitional model predictive controller used as the baseline.Unión Europea 78905Ministerio de Ciencia e Innovación PID2020-119476RB-I00Ministerio de Ciencia e Innovación IJC2018-035395-IMinisterio de Ciencia e Innovación FPU18/04476Ministerio de Ciencia e Innovación FPU20/01958ElsevierIngeniería de Sistemas y AutomáticaMinisterio de Ciencia e Innovación (MICIN). EspañaMinisterio de Ciencia e Innovación (MICIN). EspañaMinisterio de Ciencia e Innovación (MICIN). EspañaMinisterio de Ciencia e Innovación (MICIN). España2023info:eu-repo/semantics/articleinfo:eu-repo/semantics/publishedVersionapplication/pdfapplication/pdfhttps://hdl.handle.net/11441/141085https://doi.org/10.1016/j.engappai.2022.105666reponame:idUS. Depósito de Investigación de la Universidad de Sevillainstname:Universidad de Sevilla (US)InglésEngineering Applications of Artificial Intelligence, 118.78905PID2020-119476RB-I00IJC2018-035395-IFPU18/04476FPU20/01958https://www.sciencedirect.com/science/article/pii/S095219762200656Xinfo:eu-repo/semantics/openAccessoai:idus.us.es:11441/1410852026-06-17T12:51:07Z
dc.title.none.fl_str_mv A fast implementation of coalitional model predictive controllers based on machine learning: Application to solar power plants
title A fast implementation of coalitional model predictive controllers based on machine learning: Application to solar power plants
spellingShingle A fast implementation of coalitional model predictive controllers based on machine learning: Application to solar power plants
Masero Rubio, Eva
Neural networks
Artificial intelligence
Non-linear model predictive control
Coalitional control
Multi-agent systems
Solar thermal applications
title_short A fast implementation of coalitional model predictive controllers based on machine learning: Application to solar power plants
title_full A fast implementation of coalitional model predictive controllers based on machine learning: Application to solar power plants
title_fullStr A fast implementation of coalitional model predictive controllers based on machine learning: Application to solar power plants
title_full_unstemmed A fast implementation of coalitional model predictive controllers based on machine learning: Application to solar power plants
title_sort A fast implementation of coalitional model predictive controllers based on machine learning: Application to solar power plants
dc.creator.none.fl_str_mv Masero Rubio, Eva
Ruiz-Moreno, Sara
Domínguez Frejo, José Ramón
Maestre Torreblanca, José María
Camacho, Eduardo F.
author Masero Rubio, Eva
author_facet Masero Rubio, Eva
Ruiz-Moreno, Sara
Domínguez Frejo, José Ramón
Maestre Torreblanca, José María
Camacho, Eduardo F.
author_role author
author2 Ruiz-Moreno, Sara
Domínguez Frejo, José Ramón
Maestre Torreblanca, José María
Camacho, Eduardo F.
author2_role author
author
author
author
dc.contributor.none.fl_str_mv Ingeniería de Sistemas y Automática
Ministerio de Ciencia e Innovación (MICIN). España
Ministerio de Ciencia e Innovación (MICIN). España
Ministerio de Ciencia e Innovación (MICIN). España
Ministerio de Ciencia e Innovación (MICIN). España
dc.subject.none.fl_str_mv Neural networks
Artificial intelligence
Non-linear model predictive control
Coalitional control
Multi-agent systems
Solar thermal applications
topic Neural networks
Artificial intelligence
Non-linear model predictive control
Coalitional control
Multi-agent systems
Solar thermal applications
description This article proposes a real-time implementation of distributed model predictive controllers to maximize the thermal energy generated by parabolic trough collector fields. For this control strategy, we consider that each loop of the solar collector field is individually managed by a controller, which can form coalition with other controllers to attain its local goals while contributing to the overall objective. The formation of coalitions is based on a market-based mechanism in which the heat transfer fluid is traded. To relieve the computational burden online, we propose a learning-based approach that approximates optimization problems so that the controller can be applied in real time. Finally, simulations in a -loop solar collector field are used to assess the coalitional strategy based on neural networks in comparison with the coalitional model predictive control. The results show that the coalitional strategy based on neural networks provides a reduction in computing time of up to and a minimal reduction in performance compared to the coalitional model predictive controller used as the baseline.
publishDate 2023
dc.date.none.fl_str_mv 2023
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/publishedVersion
format article
status_str publishedVersion
dc.identifier.none.fl_str_mv https://hdl.handle.net/11441/141085
https://doi.org/10.1016/j.engappai.2022.105666
url https://hdl.handle.net/11441/141085
https://doi.org/10.1016/j.engappai.2022.105666
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv Engineering Applications of Artificial Intelligence, 118.
78905
PID2020-119476RB-I00
IJC2018-035395-I
FPU18/04476
FPU20/01958
https://www.sciencedirect.com/science/article/pii/S095219762200656X
dc.rights.none.fl_str_mv info:eu-repo/semantics/openAccess
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
application/pdf
dc.publisher.none.fl_str_mv Elsevier
publisher.none.fl_str_mv Elsevier
dc.source.none.fl_str_mv reponame:idUS. Depósito de Investigación de la Universidad de Sevilla
instname:Universidad de Sevilla (US)
instname_str Universidad de Sevilla (US)
reponame_str idUS. Depósito de Investigación de la Universidad de Sevilla
collection idUS. Depósito de Investigación de la Universidad de Sevilla
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869411630281916416
score 15,300719

Similares en LA Referencia