Title:	Model-free adaptive filter to mitigate actuator wear
Authors:	Fortaleza, Eugênio Libório Feitosa Gomes, Lucas M. Limaverde Filho, José Oniram de Aquino Campos, Mário C. M. M. de Longhi, Luis Gustavo S. Lima, Marcelo Lopes de Tognetti, Eduardo Stockler
metadata.dc.contributor.affiliation:	University of Brasília, Department of Mechanical Engineering, Automation and Control Group (GRACO), Brasília, Federal District, Brazil University of Brasília, Department of Mechanical Engineering, Automation and Control Group (GRACO), Brasília, Federal District, Brazil University of Brasília, Department of Mechanical Engineering, Automation and Control Group (GRACO), Brasília, Federal District, Brazil Research and Development Center (CENPES), Petróleo Brasileiro S.A. (Petrobras), Rio de Janeiro, Rio de Janeiro, Brazil PETROBRAS – UO REFAP – Alberto Pasqualini Refinery-Optimization, Brazil Research and Development Center (CENPES), Petróleo Brasileiro S.A. (Petrobras), Rio de Janeiro, Rio de Janeiro, Brazil Department of Electrical Engineering, University of Brasília, Brasília, Federal District, Brazil
Assunto::	Filtro adaptável Média móvel exponencial Aplicação industrial Controle do processo
Issue Date:	22-Feb-2022
Publisher:	Elsevier
Citation:	FORTALEZA, Eugênio L. F.; GOMES, Lucas M.; LIMAVERDE FILHO, José Oniram de A.; CAMPOS, Mário C. M. M. de; LONGHI, Luis Gustavo S.; LIMA, Marcelo Lopes de; TOGNETTI, Eduardo S. Model-free adaptive filter to mitigate actuator wear. ISA Transactions, 2022. DOI: 10.1016/j.isatra.2022.02.026. Disponível em: https://doi.org/10.1016/j.isatra.2022.02.026. Acesso em: 16 ago. 2026.
Abstract:	The present article introduces an adaptive filter of statistical basis developed for closed-loop control applications, whose goal is to reduce actuator wear while ensuring a similar control performance regarding the original closed-loop system. The main idea is to avoid the rapid change of the filtered signal when the system output has a derivative not statistically significant regarding the expected measurement noise. The adaptation law of the time constant of the filter is model-free, and the only required information is the variance of the additive noise that the measurements are subjected to. The performance of the proposed adaptive method is illustrated through a combined numerical and experimental study, in addition to its application in an operational oil plant. The results indicate that our formulation holds promise for extending the life of actuators and is easy to implement in most programmable logic controllers.
metadata.dc.description.unidade:	Faculdade de Tecnologia (FT) Departamento de Engenharia Mecânica (FT ENM) Departamento de Engenharia Elétrica (FT ENE)
DOI:	https://doi.org/10.1016/j.isatra.2022.02.026
metadata.dc.relation.publisherversion:	https://www.sciencedirect.com/science/article/pii/S0019057822000829?via%3Dihub
Appears in Collections:	Artigos publicados em periódicos e afins

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.