Rate-dependent hysteresis modeling and compensation using least squares support vector machines

Qingsong Xu; Pak Kin Wong; Yangmin Li

doi:10.1007/978-3-642-21090-7_10

Rate-dependent hysteresis modeling and compensation using least squares support vector machines

Qingsong Xu, Pak Kin Wong, Yangmin Li

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

1 Citation (Scopus)

Abstract

This paper is concentrated on the rate-dependent hysteresis modeling and compensation for a piezoelectric actuator. A least squares support vector machines (LS-SVM) model is proposed and trained by introducing the current input value and input variation rate as the input data set to formulate a one-to-one mapping. After demonstrating the effectiveness of the presented model, a LS-SVM inverse model based feedforward control combined with a PID feedback control is designed to compensate the hysteresis nonlinearity. Simulation results show that the hybrid scheme is superior to either of the stand-alone controllers, and the rate-dependent hysteresis is suppressed to a negligible level, which validate the effectiveness of the constructed controller. Owing to the simple procedure, the proposed modeling and control approaches are expected to be extended to other types of hysteretic systems as well.

Original language	English
Title of host publication	Advances in Neural Networks - 8th International Symposium on Neural Networks, ISNN 2011
Pages	85-93
Number of pages	9
Edition	PART 2
DOIs	https://doi.org/10.1007/978-3-642-21090-7_10
Publication status	Published - 6 Jun 2011
Externally published	Yes
Event	8th International Symposium on Neural Networks, ISNN 2011 - Guilin, China Duration: 29 May 2011 → 1 Jun 2011

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Number	PART 2
Volume	6676 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	8th International Symposium on Neural Networks, ISNN 2011
Country/Territory	China
City	Guilin
Period	29/05/11 → 1/06/11

Keywords

hysteresis
least squares support vector machines (LS-SVM)
motion control
Piezoelectric actuator

ASJC Scopus subject areas

Theoretical Computer Science
Computer Science(all)

More information

10.1007/978-3-642-21090-7_10

Cite this

Xu, Q., Wong, P. K., & Li, Y. (2011). Rate-dependent hysteresis modeling and compensation using least squares support vector machines. In Advances in Neural Networks - 8th International Symposium on Neural Networks, ISNN 2011 (PART 2 ed., pp. 85-93). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6676 LNCS, No. PART 2). https://doi.org/10.1007/978-3-642-21090-7_10

Xu, Qingsong ; Wong, Pak Kin ; Li, Yangmin. / Rate-dependent hysteresis modeling and compensation using least squares support vector machines. Advances in Neural Networks - 8th International Symposium on Neural Networks, ISNN 2011. PART 2. ed. 2011. pp. 85-93 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2).

@inproceedings{19d89e5b0c694c6d9a15357d29b61a02,

title = "Rate-dependent hysteresis modeling and compensation using least squares support vector machines",

abstract = "This paper is concentrated on the rate-dependent hysteresis modeling and compensation for a piezoelectric actuator. A least squares support vector machines (LS-SVM) model is proposed and trained by introducing the current input value and input variation rate as the input data set to formulate a one-to-one mapping. After demonstrating the effectiveness of the presented model, a LS-SVM inverse model based feedforward control combined with a PID feedback control is designed to compensate the hysteresis nonlinearity. Simulation results show that the hybrid scheme is superior to either of the stand-alone controllers, and the rate-dependent hysteresis is suppressed to a negligible level, which validate the effectiveness of the constructed controller. Owing to the simple procedure, the proposed modeling and control approaches are expected to be extended to other types of hysteretic systems as well.",

keywords = "hysteresis, least squares support vector machines (LS-SVM), motion control, Piezoelectric actuator",

author = "Qingsong Xu and Wong, {Pak Kin} and Yangmin Li",

year = "2011",

month = jun,

day = "6",

doi = "10.1007/978-3-642-21090-7_10",

language = "English",

isbn = "9783642210891",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

number = "PART 2",

pages = "85--93",

booktitle = "Advances in Neural Networks - 8th International Symposium on Neural Networks, ISNN 2011",

edition = "PART 2",

note = "8th International Symposium on Neural Networks, ISNN 2011 ; Conference date: 29-05-2011 Through 01-06-2011",

}

Xu, Q, Wong, PK & Li, Y 2011, Rate-dependent hysteresis modeling and compensation using least squares support vector machines. in Advances in Neural Networks - 8th International Symposium on Neural Networks, ISNN 2011. PART 2 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 2, vol. 6676 LNCS, pp. 85-93, 8th International Symposium on Neural Networks, ISNN 2011, Guilin, China, 29/05/11. https://doi.org/10.1007/978-3-642-21090-7_10

Rate-dependent hysteresis modeling and compensation using least squares support vector machines. / Xu, Qingsong; Wong, Pak Kin; Li, Yangmin.

Advances in Neural Networks - 8th International Symposium on Neural Networks, ISNN 2011. PART 2. ed. 2011. p. 85-93 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6676 LNCS, No. PART 2).

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - Rate-dependent hysteresis modeling and compensation using least squares support vector machines

AU - Xu, Qingsong

AU - Wong, Pak Kin

AU - Li, Yangmin

PY - 2011/6/6

Y1 - 2011/6/6

N2 - This paper is concentrated on the rate-dependent hysteresis modeling and compensation for a piezoelectric actuator. A least squares support vector machines (LS-SVM) model is proposed and trained by introducing the current input value and input variation rate as the input data set to formulate a one-to-one mapping. After demonstrating the effectiveness of the presented model, a LS-SVM inverse model based feedforward control combined with a PID feedback control is designed to compensate the hysteresis nonlinearity. Simulation results show that the hybrid scheme is superior to either of the stand-alone controllers, and the rate-dependent hysteresis is suppressed to a negligible level, which validate the effectiveness of the constructed controller. Owing to the simple procedure, the proposed modeling and control approaches are expected to be extended to other types of hysteretic systems as well.

AB - This paper is concentrated on the rate-dependent hysteresis modeling and compensation for a piezoelectric actuator. A least squares support vector machines (LS-SVM) model is proposed and trained by introducing the current input value and input variation rate as the input data set to formulate a one-to-one mapping. After demonstrating the effectiveness of the presented model, a LS-SVM inverse model based feedforward control combined with a PID feedback control is designed to compensate the hysteresis nonlinearity. Simulation results show that the hybrid scheme is superior to either of the stand-alone controllers, and the rate-dependent hysteresis is suppressed to a negligible level, which validate the effectiveness of the constructed controller. Owing to the simple procedure, the proposed modeling and control approaches are expected to be extended to other types of hysteretic systems as well.

KW - hysteresis

KW - least squares support vector machines (LS-SVM)

KW - motion control

KW - Piezoelectric actuator

UR - http://www.scopus.com/inward/record.url?scp=79957810475&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-21090-7_10

DO - 10.1007/978-3-642-21090-7_10

M3 - Conference article published in proceeding or book

SN - 9783642210891

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 85

EP - 93

BT - Advances in Neural Networks - 8th International Symposium on Neural Networks, ISNN 2011

T2 - 8th International Symposium on Neural Networks, ISNN 2011

Y2 - 29 May 2011 through 1 June 2011

ER -

Xu Q, Wong PK, Li Y. Rate-dependent hysteresis modeling and compensation using least squares support vector machines. In Advances in Neural Networks - 8th International Symposium on Neural Networks, ISNN 2011. PART 2 ed. 2011. p. 85-93. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 2). doi: 10.1007/978-3-642-21090-7_10

Rate-dependent hysteresis modeling and compensation using least squares support vector machines

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

More information

Other files and links

Fingerprint

Cite this