Effective optimization for fuzzy model predictive control

Stanimir Mollov, Robert Babuška, J. Abonyi, Henk B. Verbruggen

Research output: Contribution to journalArticle

111 Citations (Scopus)

Abstract

This paper addresses the optimization in fuzzy model predictive control. When the prediction model is a nonlinear fuzzy model, nonconvex, time-consuming optimization is necessary, with no guarantee of finding an optimal solution. A possible way around this problem is to linearize the fuzzy model at the current operating point and use linear predictive control (i.e., quadratic programming). For long-range predictive control, however, the influence of the linearization error may significantly deteriorate the performance. In our approach, this is remedied by linearizing the fuzzy model along the predicted input and output trajectories. One can further improve the model prediction by iteratively applying the optimized control sequence to the fuzzy model and linearizing along the so obtained simulated trajectories. Four different methods for the construction of the optimization problem are proposed, making difference between the cases when a single linear model or a set of linear models are used. By choosing an appropriate method, the user can achieve a desired tradeoff between the control performance and the computational load. The proposed techniques have been tested and evaluated using two simulated industrial benchmarks: pH control in a continuous stirred tank reactor and a high-purity distillation column.

Original languageEnglish
Pages (from-to)661-675
Number of pages15
JournalIEEE Transactions on Fuzzy Systems
Volume12
Issue number5
DOIs
Publication statusPublished - Oct 2004

Fingerprint

Model predictive control
Model Predictive Control
Fuzzy Model
Optimization
Predictive Control
Prediction Model
Linear Model
Trajectory
Distillation
Linear Control
Quadratic Programming
Reactor
Trajectories
Linearization
Nonlinear Model
Optimal Solution
Trade-offs
Quadratic programming
Distillation columns
Benchmark

ASJC Scopus subject areas

  • Control and Systems Engineering
  • Electrical and Electronic Engineering
  • Artificial Intelligence

Cite this

Effective optimization for fuzzy model predictive control. / Mollov, Stanimir; Babuška, Robert; Abonyi, J.; Verbruggen, Henk B.

In: IEEE Transactions on Fuzzy Systems, Vol. 12, No. 5, 10.2004, p. 661-675.

Research output: Contribution to journalArticle

Mollov, Stanimir ; Babuška, Robert ; Abonyi, J. ; Verbruggen, Henk B. / Effective optimization for fuzzy model predictive control. In: IEEE Transactions on Fuzzy Systems. 2004 ; Vol. 12, No. 5. pp. 661-675.
@article{a9f022ca50cd4f3aba5dd50d72c5484e,
title = "Effective optimization for fuzzy model predictive control",
abstract = "This paper addresses the optimization in fuzzy model predictive control. When the prediction model is a nonlinear fuzzy model, nonconvex, time-consuming optimization is necessary, with no guarantee of finding an optimal solution. A possible way around this problem is to linearize the fuzzy model at the current operating point and use linear predictive control (i.e., quadratic programming). For long-range predictive control, however, the influence of the linearization error may significantly deteriorate the performance. In our approach, this is remedied by linearizing the fuzzy model along the predicted input and output trajectories. One can further improve the model prediction by iteratively applying the optimized control sequence to the fuzzy model and linearizing along the so obtained simulated trajectories. Four different methods for the construction of the optimization problem are proposed, making difference between the cases when a single linear model or a set of linear models are used. By choosing an appropriate method, the user can achieve a desired tradeoff between the control performance and the computational load. The proposed techniques have been tested and evaluated using two simulated industrial benchmarks: pH control in a continuous stirred tank reactor and a high-purity distillation column.",
author = "Stanimir Mollov and Robert Babuška and J. Abonyi and Verbruggen, {Henk B.}",
year = "2004",
month = "10",
doi = "10.1109/TFUZZ.2004.834812",
language = "English",
volume = "12",
pages = "661--675",
journal = "IEEE Transactions on Fuzzy Systems",
issn = "1063-6706",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
number = "5",

}

TY - JOUR

T1 - Effective optimization for fuzzy model predictive control

AU - Mollov, Stanimir

AU - Babuška, Robert

AU - Abonyi, J.

AU - Verbruggen, Henk B.

PY - 2004/10

Y1 - 2004/10

N2 - This paper addresses the optimization in fuzzy model predictive control. When the prediction model is a nonlinear fuzzy model, nonconvex, time-consuming optimization is necessary, with no guarantee of finding an optimal solution. A possible way around this problem is to linearize the fuzzy model at the current operating point and use linear predictive control (i.e., quadratic programming). For long-range predictive control, however, the influence of the linearization error may significantly deteriorate the performance. In our approach, this is remedied by linearizing the fuzzy model along the predicted input and output trajectories. One can further improve the model prediction by iteratively applying the optimized control sequence to the fuzzy model and linearizing along the so obtained simulated trajectories. Four different methods for the construction of the optimization problem are proposed, making difference between the cases when a single linear model or a set of linear models are used. By choosing an appropriate method, the user can achieve a desired tradeoff between the control performance and the computational load. The proposed techniques have been tested and evaluated using two simulated industrial benchmarks: pH control in a continuous stirred tank reactor and a high-purity distillation column.

AB - This paper addresses the optimization in fuzzy model predictive control. When the prediction model is a nonlinear fuzzy model, nonconvex, time-consuming optimization is necessary, with no guarantee of finding an optimal solution. A possible way around this problem is to linearize the fuzzy model at the current operating point and use linear predictive control (i.e., quadratic programming). For long-range predictive control, however, the influence of the linearization error may significantly deteriorate the performance. In our approach, this is remedied by linearizing the fuzzy model along the predicted input and output trajectories. One can further improve the model prediction by iteratively applying the optimized control sequence to the fuzzy model and linearizing along the so obtained simulated trajectories. Four different methods for the construction of the optimization problem are proposed, making difference between the cases when a single linear model or a set of linear models are used. By choosing an appropriate method, the user can achieve a desired tradeoff between the control performance and the computational load. The proposed techniques have been tested and evaluated using two simulated industrial benchmarks: pH control in a continuous stirred tank reactor and a high-purity distillation column.

UR - http://www.scopus.com/inward/record.url?scp=7244236331&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=7244236331&partnerID=8YFLogxK

U2 - 10.1109/TFUZZ.2004.834812

DO - 10.1109/TFUZZ.2004.834812

M3 - Article

AN - SCOPUS:7244236331

VL - 12

SP - 661

EP - 675

JO - IEEE Transactions on Fuzzy Systems

JF - IEEE Transactions on Fuzzy Systems

SN - 1063-6706

IS - 5

ER -