A cutting plane algorithm for frequency domain specification with application to bending modes attenuation

Mohamed Abbas-Turki; Gilles Duc; Stephane Font; Benoît Clement

doi:10.1109/ACC.2005.1470746

A cutting plane algorithm for frequency domain specification with application to bending modes attenuation

Mohamed Abbas-Turki, Gilles Duc, Stephane Font, Benoît Clement

Research output: Contribution to conference › Paper › peer-review

2 Citations (Scopus)

Abstract

Bending modes attenuation is a very challenging problem, especially if the modes are close to the bandwidth, in which case applying weighting filters is useless. This work aims to present an idea based on the Cutting Plane Algorithm (CPA) to overcome such problems. The interesting point of using CPA instead of optimization under LMI (Linear Matrix Inequality) constraints is that no additional variables are introduced, which has a significant effect on the computation time. The application of Youla parameterization is considered in order to restore the linearity in the compensator parameters, thus the convexity is guaranteed and the CPA is efficiently used.

Original language	English
Pages	4748-4752
Number of pages	5
DOIs	https://doi.org/10.1109/ACC.2005.1470746
Publication status	Published - Aug 2005
Externally published	Yes
Event	2005 American Control Conference, ACC - Portland, OR, United States Duration: 8 Jun 2005 → 10 Jun 2005

Conference

Conference	2005 American Control Conference, ACC
Country/Territory	United States
City	Portland, OR
Period	8/06/05 → 10/06/05

Keywords

Cutting Plane Algorithm
KYP lemma
Youla parameterization

Access to Document

10.1109/ACC.2005.1470746Licence: In Copyright

Cite this

@conference{9c364912808f4c5db0f5aa148c33dfc5,

title = "A cutting plane algorithm for frequency domain specification with application to bending modes attenuation",

abstract = "Bending modes attenuation is a very challenging problem, especially if the modes are close to the bandwidth, in which case applying weighting filters is useless. This work aims to present an idea based on the Cutting Plane Algorithm (CPA) to overcome such problems. The interesting point of using CPA instead of optimization under LMI (Linear Matrix Inequality) constraints is that no additional variables are introduced, which has a significant effect on the computation time. The application of Youla parameterization is considered in order to restore the linearity in the compensator parameters, thus the convexity is guaranteed and the CPA is efficiently used.",

keywords = "Cutting Plane Algorithm, KYP lemma, Youla parameterization",

author = "Mohamed Abbas-Turki and Gilles Duc and Stephane Font and Beno{\^i}t Clement",

year = "2005",

month = aug,

doi = "10.1109/ACC.2005.1470746",

language = "English",

pages = "4748--4752",

note = "2005 American Control Conference, ACC ; Conference date: 08-06-2005 Through 10-06-2005",

}

TY - CONF

T1 - A cutting plane algorithm for frequency domain specification with application to bending modes attenuation

AU - Abbas-Turki, Mohamed

AU - Duc, Gilles

AU - Font, Stephane

AU - Clement, Benoît

PY - 2005/8

Y1 - 2005/8

N2 - Bending modes attenuation is a very challenging problem, especially if the modes are close to the bandwidth, in which case applying weighting filters is useless. This work aims to present an idea based on the Cutting Plane Algorithm (CPA) to overcome such problems. The interesting point of using CPA instead of optimization under LMI (Linear Matrix Inequality) constraints is that no additional variables are introduced, which has a significant effect on the computation time. The application of Youla parameterization is considered in order to restore the linearity in the compensator parameters, thus the convexity is guaranteed and the CPA is efficiently used.

AB - Bending modes attenuation is a very challenging problem, especially if the modes are close to the bandwidth, in which case applying weighting filters is useless. This work aims to present an idea based on the Cutting Plane Algorithm (CPA) to overcome such problems. The interesting point of using CPA instead of optimization under LMI (Linear Matrix Inequality) constraints is that no additional variables are introduced, which has a significant effect on the computation time. The application of Youla parameterization is considered in order to restore the linearity in the compensator parameters, thus the convexity is guaranteed and the CPA is efficiently used.

KW - Cutting Plane Algorithm

KW - KYP lemma

KW - Youla parameterization

UR - http://www.scopus.com/inward/record.url?scp=23944443690&partnerID=8YFLogxK

U2 - 10.1109/ACC.2005.1470746

DO - 10.1109/ACC.2005.1470746

M3 - Paper

AN - SCOPUS:23944443690

SP - 4748

EP - 4752

T2 - 2005 American Control Conference, ACC

Y2 - 8 June 2005 through 10 June 2005

ER -

A cutting plane algorithm for frequency domain specification with application to bending modes attenuation

Abstract

Conference

Keywords

Access to Document

Other files and links

Cite this