An efficient PID-based optimizer loop and its application in De Jong's functions minimization and quadratic regression problems

Basit Öğe Kaydı
dc.authoridDeniz, Furkan Nur/0000-0002-2524-7152
dc.authoridKoseoglu, Murat/0000-0003-3774-1083
dc.authoridAlagoz, Baris Baykant/0000-0001-5238-6433
dc.authorwosidDeniz, Furkan Nur/ABB-9604-2020
dc.authorwosidKoseoglu, Murat/ABG-8975-2020
dc.authorwosidAlagoz, Baris Baykant/ABG-8526-2020
dc.contributor.authorAlagoz, Baris Baykant
dc.contributor.authorDeniz, Furkan Nur
dc.contributor.authorKoseoglu, Murat
dc.date.accessioned2024-08-04T20:50:54Z
dc.date.available2024-08-04T20:50:54Z
dc.date.issued2022
dc.departmentİnönü Üniversitesien_US
dc.description.abstractThe Proportional-Integral-Derivative (PID) control law has been commonly used for process control in control engineering applications. However, it has potential to work as a solver in optimization problems. This study introduces a PID-based optimizer loop that is designed to solve nonlinear, unconstrained, multi-parameter optimization problems. To achieve the minimization of multi-parameter positive real objective functions by using a closed loop PID control loop, a slope sentient objective function model is suggested to allow zero-crossing of the error signal. Thus, this objective function model enhances the convergence efficiency of the PID-based optimizer loop by indicating slope direction of the objective function and operating in both positive and negative error regions. The boundedness and convergence theorems for the proposed PID optimizer loop are presented, and a theoretical background for the PID-based minimization is established. To demonstrate practical minimization performance, numerical applications of the proposed PID optimizer loops are illustrated in the solution of two fundamental optimization problems. These are the minimization of 30 parameters De Jong's functions and the solution of quadratic regression problems. Also, an experimental study is presented for the quadratic regression modeling of measurement data from a hole-drilling experiment. Optimization results reveal that the proposed PID-based optimizer system can improve convergence speed and accuracy compared to performances of fundamental nonlinear optimization techniques. (c) 2021 Elsevier B.V. All rights reserved.en_US
dc.identifier.doi10.1016/j.sysconle.2021.105090
dc.identifier.issn0167-6911
dc.identifier.issn1872-7956
dc.identifier.scopus2-s2.0-85120865084en_US
dc.identifier.scopusqualityQ1en_US
dc.identifier.urihttps://doi.org/10.1016/j.sysconle.2021.105090
dc.identifier.urihttps://hdl.handle.net/11616/100353
dc.identifier.volume159en_US
dc.identifier.wosWOS:000729423800006en_US
dc.identifier.wosqualityQ2en_US
dc.indekslendigikaynakWeb of Scienceen_US
dc.indekslendigikaynakScopusen_US
dc.language.isoenen_US
dc.publisherElsevieren_US
dc.relation.ispartofSystems & Control Lettersen_US
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanıen_US
dc.rightsinfo:eu-repo/semantics/closedAccessen_US
dc.subjectNonlinear optimizationen_US
dc.subjectPID-based optimizersen_US
dc.subjectDe Jong's functionen_US
dc.subjectQuadratic regressionen_US
dc.titleAn efficient PID-based optimizer loop and its application in De Jong's functions minimization and quadratic regression problemsen_US
dc.typeArticleen_US

Dosyalar

Koleksiyon

WoS İndeksli Yayınlar Koleksiyonu
Scopus İndeksli Yayınlar Koleksiyonu