Nonlinear controller design of active magnetic bearing systems based on polytopic quasi-LPV models

Dong Hwan Lee, Jin Bae Park, Hyun Suk Jeong, Young Hoon Joo

Research output: Contribution to journalArticle

Abstract

In this paper, a systematic procedure to design a nonlinear controller for nonlinear active magnetic bearing (AMB) systems is presented. To do this, we effectively convert the AMB system into a polytopic quasi-linear parameter varying (LPV) system, which is a representation of nonlinear state-space models and is described by the convex combination of a set of precisely known vertices. Unlike the existing quasi-LPV systems, the nonlinear weighting functions, which construct the polytopic quasi-LPV model of the AMB system by connecting the vertices, include not only state variables but also the input ones. This allows us to treat the input nonlinearity effectively. By means of the derived polytopic quasi-LPV model and linear matrix inequality (LMI) conditions, nonlinear controller that stabilizes the AMB system is obtained. The effectiveness of the proposed controller design methodology is finally demonstrated through numerical simulations.

Original languageEnglish
Pages (from-to)797-802
Number of pages6
JournalTransactions of the Korean Institute of Electrical Engineers
Volume59
Issue number4
Publication statusPublished - 2010 Apr 1

Fingerprint

Magnetic bearings
Controllers
Linear matrix inequalities
Computer simulation

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering

Cite this

@article{037e851163694a5aa4c9f5af78461dfd,
title = "Nonlinear controller design of active magnetic bearing systems based on polytopic quasi-LPV models",
abstract = "In this paper, a systematic procedure to design a nonlinear controller for nonlinear active magnetic bearing (AMB) systems is presented. To do this, we effectively convert the AMB system into a polytopic quasi-linear parameter varying (LPV) system, which is a representation of nonlinear state-space models and is described by the convex combination of a set of precisely known vertices. Unlike the existing quasi-LPV systems, the nonlinear weighting functions, which construct the polytopic quasi-LPV model of the AMB system by connecting the vertices, include not only state variables but also the input ones. This allows us to treat the input nonlinearity effectively. By means of the derived polytopic quasi-LPV model and linear matrix inequality (LMI) conditions, nonlinear controller that stabilizes the AMB system is obtained. The effectiveness of the proposed controller design methodology is finally demonstrated through numerical simulations.",
author = "Lee, {Dong Hwan} and Park, {Jin Bae} and Jeong, {Hyun Suk} and Joo, {Young Hoon}",
year = "2010",
month = "4",
day = "1",
language = "English",
volume = "59",
pages = "797--802",
journal = "Transactions of the Korean Institute of Electrical Engineers",
issn = "1975-8359",
publisher = "Korean Institute of Electrical Engineers",
number = "4",

}

Nonlinear controller design of active magnetic bearing systems based on polytopic quasi-LPV models. / Lee, Dong Hwan; Park, Jin Bae; Jeong, Hyun Suk; Joo, Young Hoon.

In: Transactions of the Korean Institute of Electrical Engineers, Vol. 59, No. 4, 01.04.2010, p. 797-802.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Nonlinear controller design of active magnetic bearing systems based on polytopic quasi-LPV models

AU - Lee, Dong Hwan

AU - Park, Jin Bae

AU - Jeong, Hyun Suk

AU - Joo, Young Hoon

PY - 2010/4/1

Y1 - 2010/4/1

N2 - In this paper, a systematic procedure to design a nonlinear controller for nonlinear active magnetic bearing (AMB) systems is presented. To do this, we effectively convert the AMB system into a polytopic quasi-linear parameter varying (LPV) system, which is a representation of nonlinear state-space models and is described by the convex combination of a set of precisely known vertices. Unlike the existing quasi-LPV systems, the nonlinear weighting functions, which construct the polytopic quasi-LPV model of the AMB system by connecting the vertices, include not only state variables but also the input ones. This allows us to treat the input nonlinearity effectively. By means of the derived polytopic quasi-LPV model and linear matrix inequality (LMI) conditions, nonlinear controller that stabilizes the AMB system is obtained. The effectiveness of the proposed controller design methodology is finally demonstrated through numerical simulations.

AB - In this paper, a systematic procedure to design a nonlinear controller for nonlinear active magnetic bearing (AMB) systems is presented. To do this, we effectively convert the AMB system into a polytopic quasi-linear parameter varying (LPV) system, which is a representation of nonlinear state-space models and is described by the convex combination of a set of precisely known vertices. Unlike the existing quasi-LPV systems, the nonlinear weighting functions, which construct the polytopic quasi-LPV model of the AMB system by connecting the vertices, include not only state variables but also the input ones. This allows us to treat the input nonlinearity effectively. By means of the derived polytopic quasi-LPV model and linear matrix inequality (LMI) conditions, nonlinear controller that stabilizes the AMB system is obtained. The effectiveness of the proposed controller design methodology is finally demonstrated through numerical simulations.

UR - http://www.scopus.com/inward/record.url?scp=77951216474&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77951216474&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:77951216474

VL - 59

SP - 797

EP - 802

JO - Transactions of the Korean Institute of Electrical Engineers

JF - Transactions of the Korean Institute of Electrical Engineers

SN - 1975-8359

IS - 4

ER -