Gray-box nonlinear system identification using polynomial NARMAX models
Autor(a) principal: | |
---|---|
Data de Publicação: | 2019 |
Tipo de documento: | Dissertação |
Idioma: | eng |
Título da fonte: | Repositório Institucional da UFRJ |
Texto Completo: | http://hdl.handle.net/11422/13603 |
Resumo: | The usage of a collection of linear models to describe a nonlinear system has many disadvantages. In order to overcome these disadvantages, nonlinear models have been improved. The nonlinear model used in this work is the Nonlinear AutoRegressive Moving Average models with eXogenous inputs (NARMAX) of polynomial type. This type of model is linear on the parameters and accounts, in the model, for the existent noise, that is inherent of a measurement on a industrial plant. Broadly, there are two types of identification: the black-box identification, which is a typical input-output method, i.e., only requires data in order to identify the process; and the gray-box identification, which requires some system information, besides data. In the present work, a gray-box identification is compared with the black-box one for optimization and control purposes. The identification is performed using the Orthogonal Least Square algorithm and validation is made using k-stepahead cross-validation method. Dynamic real-time optimization was set based on both first principle models and estimated models, and compared, in order to evaluate improvement on the application of identified nonlinear models. The gray-box identification was more representative in relation to the nonlinearity of the system. The application in optimization and control generated instability of the algorithm. It can be due to the fact that the optimization algorithm used in dynamic real-time optimization had the same value for control horizon and prediction horizon. Despite the oscillations of one case study, the gray-box identification algorithm showed its capacity to improve the model. |
id |
UFRJ_53c3fc96689fa43995301d445b2f3c44 |
---|---|
oai_identifier_str |
oai:pantheon.ufrj.br:11422/13603 |
network_acronym_str |
UFRJ |
network_name_str |
Repositório Institucional da UFRJ |
repository_id_str |
|
spelling |
Gray-box nonlinear system identification using polynomial NARMAX modelsIdentificação caixa-cinza de sistemas não lineares usando modelos NARMAX polinomiaisNARMAX modelNonlinear systemsGray-boxCNPQ::ENGENHARIAS::ENGENHARIA QUIMICAThe usage of a collection of linear models to describe a nonlinear system has many disadvantages. In order to overcome these disadvantages, nonlinear models have been improved. The nonlinear model used in this work is the Nonlinear AutoRegressive Moving Average models with eXogenous inputs (NARMAX) of polynomial type. This type of model is linear on the parameters and accounts, in the model, for the existent noise, that is inherent of a measurement on a industrial plant. Broadly, there are two types of identification: the black-box identification, which is a typical input-output method, i.e., only requires data in order to identify the process; and the gray-box identification, which requires some system information, besides data. In the present work, a gray-box identification is compared with the black-box one for optimization and control purposes. The identification is performed using the Orthogonal Least Square algorithm and validation is made using k-stepahead cross-validation method. Dynamic real-time optimization was set based on both first principle models and estimated models, and compared, in order to evaluate improvement on the application of identified nonlinear models. The gray-box identification was more representative in relation to the nonlinearity of the system. The application in optimization and control generated instability of the algorithm. It can be due to the fact that the optimization algorithm used in dynamic real-time optimization had the same value for control horizon and prediction horizon. Despite the oscillations of one case study, the gray-box identification algorithm showed its capacity to improve the model.O uso de um conjunto de modelos lineares para descrever um sistema não-linear tem muitas desvantagens. Para superar essas desvantagens, modelos não-lineares foram aprimorados. O modelo não-linear usado neste trabalho é o modelo de média móvel auto-regressiva não-linear com entradas exógenas, do inglês ”Nonlinear AutoRegressive Moving Average models with eXogenous inputs” (NARMAX) do tipo polinomial. Esse tipo de modelo é linear nos parâmetros e considera, no modelo, o ruído, inerente a uma medição em uma planta industrial. Em geral, existem dois tipos de identificação: a identificação caixa-preta, que é um método típico de entrada e saída, ou seja, requer apenas dados para identificar o processo; e a identificação da caixa-cinza, que requer algumas informações sobre o sistema, além de dados. No presente trabalho, um tipo caixa-cinza é comparado com o tipo caixapreta para fins de otimização e controle. A identificação é realizada usando o algoritmo de mínimos quadrados ortogonais e método de validação cruzada de k passos a frente. A otimização dinâmica em tempo real foi definida com base no modelo fenomenológico e em modelos estimados, e comparadas, para avaliar a melhoria na aplicação de modelos não lineares identificados. A identificação do tipo caixa-cinza se mostrou mais representativa em relação à não linearidade do sistema. A aplicação em otimização e controle gerou instabilidade do algoritmo. Isso pode ser devido ao fato de que o algoritmo de otimização usado na otimização dinâmica em tempo real tinha o mesmo valor para horizonte de controle e horizonte de predição. Apesar das oscilações de um estudo de caso, o algoritmo de identificação caixa-cinza mostrou sua capacidade de melhorar o modelo.Universidade Federal do Rio de JaneiroBrasilInstituto Alberto Luiz Coimbra de Pós-Graduação e Pesquisa de EngenhariaPrograma de Pós-Graduação em Engenharia QuímicaUFRJSecchi, Argimiro Resendehttp://lattes.cnpq.br/3710340061939187http://lattes.cnpq.br/8929685017853058Souza Júnior, Maurício Bezerra dehttp://lattes.cnpq.br/4530858702685674Capron, Bruno Didier OliverMeleiro, Luiz Augusto da CruzSantos, Allyne Machado dos2021-02-02T23:31:59Z2023-12-21T03:07:24Z2019-02info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/masterThesishttp://hdl.handle.net/11422/13603enginfo:eu-repo/semantics/openAccessreponame:Repositório Institucional da UFRJinstname:Universidade Federal do Rio de Janeiro (UFRJ)instacron:UFRJ2023-12-21T03:07:24Zoai:pantheon.ufrj.br:11422/13603Repositório InstitucionalPUBhttp://www.pantheon.ufrj.br/oai/requestpantheon@sibi.ufrj.bropendoar:2023-12-21T03:07:24Repositório Institucional da UFRJ - Universidade Federal do Rio de Janeiro (UFRJ)false |
dc.title.none.fl_str_mv |
Gray-box nonlinear system identification using polynomial NARMAX models Identificação caixa-cinza de sistemas não lineares usando modelos NARMAX polinomiais |
title |
Gray-box nonlinear system identification using polynomial NARMAX models |
spellingShingle |
Gray-box nonlinear system identification using polynomial NARMAX models Santos, Allyne Machado dos NARMAX model Nonlinear systems Gray-box CNPQ::ENGENHARIAS::ENGENHARIA QUIMICA |
title_short |
Gray-box nonlinear system identification using polynomial NARMAX models |
title_full |
Gray-box nonlinear system identification using polynomial NARMAX models |
title_fullStr |
Gray-box nonlinear system identification using polynomial NARMAX models |
title_full_unstemmed |
Gray-box nonlinear system identification using polynomial NARMAX models |
title_sort |
Gray-box nonlinear system identification using polynomial NARMAX models |
author |
Santos, Allyne Machado dos |
author_facet |
Santos, Allyne Machado dos |
author_role |
author |
dc.contributor.none.fl_str_mv |
Secchi, Argimiro Resende http://lattes.cnpq.br/3710340061939187 http://lattes.cnpq.br/8929685017853058 Souza Júnior, Maurício Bezerra de http://lattes.cnpq.br/4530858702685674 Capron, Bruno Didier Oliver Meleiro, Luiz Augusto da Cruz |
dc.contributor.author.fl_str_mv |
Santos, Allyne Machado dos |
dc.subject.por.fl_str_mv |
NARMAX model Nonlinear systems Gray-box CNPQ::ENGENHARIAS::ENGENHARIA QUIMICA |
topic |
NARMAX model Nonlinear systems Gray-box CNPQ::ENGENHARIAS::ENGENHARIA QUIMICA |
description |
The usage of a collection of linear models to describe a nonlinear system has many disadvantages. In order to overcome these disadvantages, nonlinear models have been improved. The nonlinear model used in this work is the Nonlinear AutoRegressive Moving Average models with eXogenous inputs (NARMAX) of polynomial type. This type of model is linear on the parameters and accounts, in the model, for the existent noise, that is inherent of a measurement on a industrial plant. Broadly, there are two types of identification: the black-box identification, which is a typical input-output method, i.e., only requires data in order to identify the process; and the gray-box identification, which requires some system information, besides data. In the present work, a gray-box identification is compared with the black-box one for optimization and control purposes. The identification is performed using the Orthogonal Least Square algorithm and validation is made using k-stepahead cross-validation method. Dynamic real-time optimization was set based on both first principle models and estimated models, and compared, in order to evaluate improvement on the application of identified nonlinear models. The gray-box identification was more representative in relation to the nonlinearity of the system. The application in optimization and control generated instability of the algorithm. It can be due to the fact that the optimization algorithm used in dynamic real-time optimization had the same value for control horizon and prediction horizon. Despite the oscillations of one case study, the gray-box identification algorithm showed its capacity to improve the model. |
publishDate |
2019 |
dc.date.none.fl_str_mv |
2019-02 2021-02-02T23:31:59Z 2023-12-21T03:07:24Z |
dc.type.status.fl_str_mv |
info:eu-repo/semantics/publishedVersion |
dc.type.driver.fl_str_mv |
info:eu-repo/semantics/masterThesis |
format |
masterThesis |
status_str |
publishedVersion |
dc.identifier.uri.fl_str_mv |
http://hdl.handle.net/11422/13603 |
url |
http://hdl.handle.net/11422/13603 |
dc.language.iso.fl_str_mv |
eng |
language |
eng |
dc.rights.driver.fl_str_mv |
info:eu-repo/semantics/openAccess |
eu_rights_str_mv |
openAccess |
dc.publisher.none.fl_str_mv |
Universidade Federal do Rio de Janeiro Brasil Instituto Alberto Luiz Coimbra de Pós-Graduação e Pesquisa de Engenharia Programa de Pós-Graduação em Engenharia Química UFRJ |
publisher.none.fl_str_mv |
Universidade Federal do Rio de Janeiro Brasil Instituto Alberto Luiz Coimbra de Pós-Graduação e Pesquisa de Engenharia Programa de Pós-Graduação em Engenharia Química UFRJ |
dc.source.none.fl_str_mv |
reponame:Repositório Institucional da UFRJ instname:Universidade Federal do Rio de Janeiro (UFRJ) instacron:UFRJ |
instname_str |
Universidade Federal do Rio de Janeiro (UFRJ) |
instacron_str |
UFRJ |
institution |
UFRJ |
reponame_str |
Repositório Institucional da UFRJ |
collection |
Repositório Institucional da UFRJ |
repository.name.fl_str_mv |
Repositório Institucional da UFRJ - Universidade Federal do Rio de Janeiro (UFRJ) |
repository.mail.fl_str_mv |
pantheon@sibi.ufrj.br |
_version_ |
1815456012890537984 |