LS-SVR based nonlinear adaptive direct generalized predictive control

Chuntao Cai; Jianhua Li; Qunli Xia

doi:10.1049/cp.2012.1223

LS-SVR based nonlinear adaptive direct generalized predictive control

Chuntao Cai, Jianhua Li, Qunli Xia

School of Aerospace Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Citation (Scopus)

Abstract

Generalized Predictive Control (GPC) has a strong ability of overcoming load disturbance, random noise and delay change, and the selected model has less parameter. But it also has some problems such as big amount of calculation and no consideration on both rapidity and over modulation. An adaptive direct GPC method is proposed based on LSSVR and tracking error. This method uses LS-SVR method to design predictive controller, and uses a modified projection algorithm based on tracking error to adjust the weight of LS-SVR adaptively, in order to avoid the calculating of the inverse matrix. The result indicates that the method is not only very effective but also reduces the calculation amount.

Original language	English
Title of host publication	International Conference on Automatic Control and Artificial Intelligence, ACAI 2012
Pages	1324-1326
Number of pages	3
Edition	598 CP
DOIs	https://doi.org/10.1049/cp.2012.1223
Publication status	Published - 2012
Event	International Conference on Automatic Control and Artificial Intelligence, ACAI 2012 - Xiamen, China Duration: 3 Mar 2012 → 5 Mar 2012

Publication series

Name	IET Conference Publications
Number	598 CP
Volume	2012

Conference

Conference	International Conference on Automatic Control and Artificial Intelligence, ACAI 2012
Country/Territory	China
City	Xiamen
Period	3/03/12 → 5/03/12

Keywords

Generalized predictive control
LS-SVR
Nonlinear system
Selfadaptive

Access to Document

10.1049/cp.2012.1223

Cite this

@inproceedings{0b22c1a400f64173830f429a418b8f59,

title = "LS-SVR based nonlinear adaptive direct generalized predictive control",

abstract = "Generalized Predictive Control (GPC) has a strong ability of overcoming load disturbance, random noise and delay change, and the selected model has less parameter. But it also has some problems such as big amount of calculation and no consideration on both rapidity and over modulation. An adaptive direct GPC method is proposed based on LSSVR and tracking error. This method uses LS-SVR method to design predictive controller, and uses a modified projection algorithm based on tracking error to adjust the weight of LS-SVR adaptively, in order to avoid the calculating of the inverse matrix. The result indicates that the method is not only very effective but also reduces the calculation amount.",

keywords = "Generalized predictive control, LS-SVR, Nonlinear system, Selfadaptive",

author = "Chuntao Cai and Jianhua Li and Qunli Xia",

year = "2012",

doi = "10.1049/cp.2012.1223",

language = "English",

isbn = "9781849195379",

series = "IET Conference Publications",

number = "598 CP",

pages = "1324--1326",

booktitle = "International Conference on Automatic Control and Artificial Intelligence, ACAI 2012",

edition = "598 CP",

note = "International Conference on Automatic Control and Artificial Intelligence, ACAI 2012 ; Conference date: 03-03-2012 Through 05-03-2012",

}

Cai, C, Li, J & Xia, Q 2012, LS-SVR based nonlinear adaptive direct generalized predictive control. in International Conference on Automatic Control and Artificial Intelligence, ACAI 2012. 598 CP edn, IET Conference Publications, no. 598 CP, vol. 2012, pp. 1324-1326, International Conference on Automatic Control and Artificial Intelligence, ACAI 2012, Xiamen, China, 3/03/12. https://doi.org/10.1049/cp.2012.1223

LS-SVR based nonlinear adaptive direct generalized predictive control. / Cai, Chuntao; Li, Jianhua; Xia, Qunli.
International Conference on Automatic Control and Artificial Intelligence, ACAI 2012. 598 CP. ed. 2012. p. 1324-1326 (IET Conference Publications; Vol. 2012, No. 598 CP).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - LS-SVR based nonlinear adaptive direct generalized predictive control

AU - Cai, Chuntao

AU - Li, Jianhua

AU - Xia, Qunli

PY - 2012

Y1 - 2012

N2 - Generalized Predictive Control (GPC) has a strong ability of overcoming load disturbance, random noise and delay change, and the selected model has less parameter. But it also has some problems such as big amount of calculation and no consideration on both rapidity and over modulation. An adaptive direct GPC method is proposed based on LSSVR and tracking error. This method uses LS-SVR method to design predictive controller, and uses a modified projection algorithm based on tracking error to adjust the weight of LS-SVR adaptively, in order to avoid the calculating of the inverse matrix. The result indicates that the method is not only very effective but also reduces the calculation amount.

AB - Generalized Predictive Control (GPC) has a strong ability of overcoming load disturbance, random noise and delay change, and the selected model has less parameter. But it also has some problems such as big amount of calculation and no consideration on both rapidity and over modulation. An adaptive direct GPC method is proposed based on LSSVR and tracking error. This method uses LS-SVR method to design predictive controller, and uses a modified projection algorithm based on tracking error to adjust the weight of LS-SVR adaptively, in order to avoid the calculating of the inverse matrix. The result indicates that the method is not only very effective but also reduces the calculation amount.

KW - Generalized predictive control

KW - LS-SVR

KW - Nonlinear system

KW - Selfadaptive

UR - http://www.scopus.com/inward/record.url?scp=84877788654&partnerID=8YFLogxK

U2 - 10.1049/cp.2012.1223

DO - 10.1049/cp.2012.1223

M3 - Conference contribution

AN - SCOPUS:84877788654

SN - 9781849195379

T3 - IET Conference Publications

SP - 1324

EP - 1326

BT - International Conference on Automatic Control and Artificial Intelligence, ACAI 2012

T2 - International Conference on Automatic Control and Artificial Intelligence, ACAI 2012

Y2 - 3 March 2012 through 5 March 2012

ER -

LS-SVR based nonlinear adaptive direct generalized predictive control

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this