Fuzzy-neural predictive control using fast optimisation polices

Margarita Terziyska, Yancho Todorov

Research output: Contribution to journalArticleScientificpeer-review

Abstract

This paper describes the development of fast optimisation polices based on Newtonian approaches, as effective algorithms to solve the on-line optimisation task, during the operation of a predictive controller. To simplify the calculation of the control actions, an iterative solutions based on Newton-Raphson and Levenberg-Marquardt approaches, are proposed. To avoid the computational load related to Hessian inversion, a simple Gaussian elimination in a form of matrix decomposition is applied. As plant response predictor, a Takagi-Sugeno fuzzy-neural network, with global and local (after the rules layer) recurrent nodes, is used. The efficiency of the proposed optimisation strategies is demonstrated by simulation experiments in MATLAB environment to control a continuous stirred tank reactor.

Original languageEnglish
Pages (from-to)136-144
Number of pages9
JournalInternational Journal of Reasoning-based Intelligent Systems
Volume6
Issue number3-4
DOIs
Publication statusPublished - 1 Jan 2014
MoE publication typeA1 Journal article-refereed

Fingerprint

Law enforcement
Fuzzy neural networks
MATLAB
Decomposition
Controllers
Experiments

Keywords

  • Gradient descent
  • Newton method
  • Nonlinear control
  • Optimisation
  • Predictive control
  • Takagi-Sugeno fuzzy-neural networks

Cite this

@article{168b17aaf79344f0860e1ee5ef8e54a3,
title = "Fuzzy-neural predictive control using fast optimisation polices",
abstract = "This paper describes the development of fast optimisation polices based on Newtonian approaches, as effective algorithms to solve the on-line optimisation task, during the operation of a predictive controller. To simplify the calculation of the control actions, an iterative solutions based on Newton-Raphson and Levenberg-Marquardt approaches, are proposed. To avoid the computational load related to Hessian inversion, a simple Gaussian elimination in a form of matrix decomposition is applied. As plant response predictor, a Takagi-Sugeno fuzzy-neural network, with global and local (after the rules layer) recurrent nodes, is used. The efficiency of the proposed optimisation strategies is demonstrated by simulation experiments in MATLAB environment to control a continuous stirred tank reactor.",
keywords = "Gradient descent, Newton method, Nonlinear control, Optimisation, Predictive control, Takagi-Sugeno fuzzy-neural networks",
author = "Margarita Terziyska and Yancho Todorov",
year = "2014",
month = "1",
day = "1",
doi = "10.1504/IJRIS.2014.066250",
language = "English",
volume = "6",
pages = "136--144",
journal = "International Journal of Reasoning-based Intelligent Systems",
issn = "1755-0556",
publisher = "Inderscience Publishers",
number = "3-4",

}

Fuzzy-neural predictive control using fast optimisation polices. / Terziyska, Margarita; Todorov, Yancho.

In: International Journal of Reasoning-based Intelligent Systems, Vol. 6, No. 3-4, 01.01.2014, p. 136-144.

Research output: Contribution to journalArticleScientificpeer-review

TY - JOUR

T1 - Fuzzy-neural predictive control using fast optimisation polices

AU - Terziyska, Margarita

AU - Todorov, Yancho

PY - 2014/1/1

Y1 - 2014/1/1

N2 - This paper describes the development of fast optimisation polices based on Newtonian approaches, as effective algorithms to solve the on-line optimisation task, during the operation of a predictive controller. To simplify the calculation of the control actions, an iterative solutions based on Newton-Raphson and Levenberg-Marquardt approaches, are proposed. To avoid the computational load related to Hessian inversion, a simple Gaussian elimination in a form of matrix decomposition is applied. As plant response predictor, a Takagi-Sugeno fuzzy-neural network, with global and local (after the rules layer) recurrent nodes, is used. The efficiency of the proposed optimisation strategies is demonstrated by simulation experiments in MATLAB environment to control a continuous stirred tank reactor.

AB - This paper describes the development of fast optimisation polices based on Newtonian approaches, as effective algorithms to solve the on-line optimisation task, during the operation of a predictive controller. To simplify the calculation of the control actions, an iterative solutions based on Newton-Raphson and Levenberg-Marquardt approaches, are proposed. To avoid the computational load related to Hessian inversion, a simple Gaussian elimination in a form of matrix decomposition is applied. As plant response predictor, a Takagi-Sugeno fuzzy-neural network, with global and local (after the rules layer) recurrent nodes, is used. The efficiency of the proposed optimisation strategies is demonstrated by simulation experiments in MATLAB environment to control a continuous stirred tank reactor.

KW - Gradient descent

KW - Newton method

KW - Nonlinear control

KW - Optimisation

KW - Predictive control

KW - Takagi-Sugeno fuzzy-neural networks

UR - http://www.scopus.com/inward/record.url?scp=84918519878&partnerID=8YFLogxK

U2 - 10.1504/IJRIS.2014.066250

DO - 10.1504/IJRIS.2014.066250

M3 - Article

AN - SCOPUS:84918519878

VL - 6

SP - 136

EP - 144

JO - International Journal of Reasoning-based Intelligent Systems

JF - International Journal of Reasoning-based Intelligent Systems

SN - 1755-0556

IS - 3-4

ER -