Robust predictive feedback control for constrained systems

Leonardo Giovanini, Michael Grimble

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

A new method for the design of predictive controllers for SISO systems is presented. The proposed technique allows uncertainties and constraints to be concluded in the design of the control law. The goal is to design, at each sample instant, a predictive feedback control law that minimizes a performance measure and guarantees of constraints are satisfied for a set of models that describes the system to be controlled. The predictive controller consists of a finite horizon parametric-optimization problem with an additional constraint over the manipulated variable behavior. This is an end-constraint based approach that ensures the exponential stability of the closed-loop system. The inclusion of this additional constraint, in the on-line optimization algorithm, enables robust stability properties to be demonstrated for the closed-loop system. This is the case even though constraints and disturbances are present. Finally, simulation results are presented using a nonlinear continuous stirred tank reactor model.
LanguageEnglish
Pages407-422
Number of pages16
JournalInternational Journal of Control, Automation and Systems
Volume2
Issue number4
Publication statusPublished - 2004

Fingerprint

Feedback control
Closed loop systems
Controllers
Asymptotic stability
Robust stability
Uncertainty

Keywords

  • predictive control
  • parametric optimization
  • multi-objective optimization
  • control systems
  • predictive feedback control

Cite this

@article{2a9f29867f654f1f894ba812d0fce32f,
title = "Robust predictive feedback control for constrained systems",
abstract = "A new method for the design of predictive controllers for SISO systems is presented. The proposed technique allows uncertainties and constraints to be concluded in the design of the control law. The goal is to design, at each sample instant, a predictive feedback control law that minimizes a performance measure and guarantees of constraints are satisfied for a set of models that describes the system to be controlled. The predictive controller consists of a finite horizon parametric-optimization problem with an additional constraint over the manipulated variable behavior. This is an end-constraint based approach that ensures the exponential stability of the closed-loop system. The inclusion of this additional constraint, in the on-line optimization algorithm, enables robust stability properties to be demonstrated for the closed-loop system. This is the case even though constraints and disturbances are present. Finally, simulation results are presented using a nonlinear continuous stirred tank reactor model.",
keywords = "predictive control, parametric optimization, multi-objective optimization, control systems, predictive feedback control",
author = "Leonardo Giovanini and Michael Grimble",
year = "2004",
language = "English",
volume = "2",
pages = "407--422",
journal = "International Journal of Control, Automation and Systems",
issn = "1598-6446",
number = "4",

}

Robust predictive feedback control for constrained systems. / Giovanini, Leonardo; Grimble, Michael.

In: International Journal of Control, Automation and Systems, Vol. 2, No. 4, 2004, p. 407-422.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Robust predictive feedback control for constrained systems

AU - Giovanini, Leonardo

AU - Grimble, Michael

PY - 2004

Y1 - 2004

N2 - A new method for the design of predictive controllers for SISO systems is presented. The proposed technique allows uncertainties and constraints to be concluded in the design of the control law. The goal is to design, at each sample instant, a predictive feedback control law that minimizes a performance measure and guarantees of constraints are satisfied for a set of models that describes the system to be controlled. The predictive controller consists of a finite horizon parametric-optimization problem with an additional constraint over the manipulated variable behavior. This is an end-constraint based approach that ensures the exponential stability of the closed-loop system. The inclusion of this additional constraint, in the on-line optimization algorithm, enables robust stability properties to be demonstrated for the closed-loop system. This is the case even though constraints and disturbances are present. Finally, simulation results are presented using a nonlinear continuous stirred tank reactor model.

AB - A new method for the design of predictive controllers for SISO systems is presented. The proposed technique allows uncertainties and constraints to be concluded in the design of the control law. The goal is to design, at each sample instant, a predictive feedback control law that minimizes a performance measure and guarantees of constraints are satisfied for a set of models that describes the system to be controlled. The predictive controller consists of a finite horizon parametric-optimization problem with an additional constraint over the manipulated variable behavior. This is an end-constraint based approach that ensures the exponential stability of the closed-loop system. The inclusion of this additional constraint, in the on-line optimization algorithm, enables robust stability properties to be demonstrated for the closed-loop system. This is the case even though constraints and disturbances are present. Finally, simulation results are presented using a nonlinear continuous stirred tank reactor model.

KW - predictive control

KW - parametric optimization

KW - multi-objective optimization

KW - control systems

KW - predictive feedback control

UR - http://www.ijcas.org/

M3 - Article

VL - 2

SP - 407

EP - 422

JO - International Journal of Control, Automation and Systems

T2 - International Journal of Control, Automation and Systems

JF - International Journal of Control, Automation and Systems

SN - 1598-6446

IS - 4

ER -