Kriging metamodel based statistical global optimization technique

Tae Hee Lee, S. K. Cho

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Global optimization techniques have gained much attention in the design of industrial products because they can find not only systematically the best design solution within the design space, but also provide engineers with multiple alternatives. However, the considerable computational burden of global optimization has been still a challenging problem. In this paper, statistical global optimization technique using the kriging metamodel is proposed in order to resolve the difficulty. Because the proposed method explores a global optimum by means of kriging metamodel that can evaluate inexpensively and quickly responses, computational cost of global optimization can be remarkably reduced. For mathematical problems with strong nonlinearity, it is shown that the proposed method can search for a global optimum accurately as well as efficiently. Finally, global optimization of bogie frame is provided as a practical example.

Original languageEnglish
Title of host publicationProceedings of the 10th International Conference on Computational Structures Technology, CST 2010
PublisherCivil-Comp Press
Volume93
ISBN (Print)9781905088386
StatePublished - 2010 Jan 1
Event10th International Conference on Computational Structures Technology, CST 2010 - Valencia, Spain
Duration: 2010 Sep 142010 Sep 17

Other

Other10th International Conference on Computational Structures Technology, CST 2010
CountrySpain
CityValencia
Period10/09/1410/09/17

Fingerprint

Global optimization
Engineers
Costs

Keywords

  • Global optimization technique
  • Kriging metamodel
  • Metamodel-based design optimization
  • Non-gradient optimization
  • Statistical global optimization

Cite this

Lee, T. H., & Cho, S. K. (2010). Kriging metamodel based statistical global optimization technique. In Proceedings of the 10th International Conference on Computational Structures Technology, CST 2010 (Vol. 93). Civil-Comp Press.
Lee, Tae Hee ; Cho, S. K. / Kriging metamodel based statistical global optimization technique. Proceedings of the 10th International Conference on Computational Structures Technology, CST 2010. Vol. 93 Civil-Comp Press, 2010.
@inproceedings{4a1e9ffb73b74cef91b4f7c9025c1b03,
title = "Kriging metamodel based statistical global optimization technique",
abstract = "Global optimization techniques have gained much attention in the design of industrial products because they can find not only systematically the best design solution within the design space, but also provide engineers with multiple alternatives. However, the considerable computational burden of global optimization has been still a challenging problem. In this paper, statistical global optimization technique using the kriging metamodel is proposed in order to resolve the difficulty. Because the proposed method explores a global optimum by means of kriging metamodel that can evaluate inexpensively and quickly responses, computational cost of global optimization can be remarkably reduced. For mathematical problems with strong nonlinearity, it is shown that the proposed method can search for a global optimum accurately as well as efficiently. Finally, global optimization of bogie frame is provided as a practical example.",
keywords = "Global optimization technique, Kriging metamodel, Metamodel-based design optimization, Non-gradient optimization, Statistical global optimization",
author = "Lee, {Tae Hee} and Cho, {S. K.}",
year = "2010",
month = "1",
day = "1",
language = "English",
isbn = "9781905088386",
volume = "93",
booktitle = "Proceedings of the 10th International Conference on Computational Structures Technology, CST 2010",
publisher = "Civil-Comp Press",

}

Lee, TH & Cho, SK 2010, Kriging metamodel based statistical global optimization technique. in Proceedings of the 10th International Conference on Computational Structures Technology, CST 2010. vol. 93, Civil-Comp Press, 10th International Conference on Computational Structures Technology, CST 2010, Valencia, Spain, 10/09/14.

Kriging metamodel based statistical global optimization technique. / Lee, Tae Hee; Cho, S. K.

Proceedings of the 10th International Conference on Computational Structures Technology, CST 2010. Vol. 93 Civil-Comp Press, 2010.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Kriging metamodel based statistical global optimization technique

AU - Lee, Tae Hee

AU - Cho, S. K.

PY - 2010/1/1

Y1 - 2010/1/1

N2 - Global optimization techniques have gained much attention in the design of industrial products because they can find not only systematically the best design solution within the design space, but also provide engineers with multiple alternatives. However, the considerable computational burden of global optimization has been still a challenging problem. In this paper, statistical global optimization technique using the kriging metamodel is proposed in order to resolve the difficulty. Because the proposed method explores a global optimum by means of kriging metamodel that can evaluate inexpensively and quickly responses, computational cost of global optimization can be remarkably reduced. For mathematical problems with strong nonlinearity, it is shown that the proposed method can search for a global optimum accurately as well as efficiently. Finally, global optimization of bogie frame is provided as a practical example.

AB - Global optimization techniques have gained much attention in the design of industrial products because they can find not only systematically the best design solution within the design space, but also provide engineers with multiple alternatives. However, the considerable computational burden of global optimization has been still a challenging problem. In this paper, statistical global optimization technique using the kriging metamodel is proposed in order to resolve the difficulty. Because the proposed method explores a global optimum by means of kriging metamodel that can evaluate inexpensively and quickly responses, computational cost of global optimization can be remarkably reduced. For mathematical problems with strong nonlinearity, it is shown that the proposed method can search for a global optimum accurately as well as efficiently. Finally, global optimization of bogie frame is provided as a practical example.

KW - Global optimization technique

KW - Kriging metamodel

KW - Metamodel-based design optimization

KW - Non-gradient optimization

KW - Statistical global optimization

UR - http://www.scopus.com/inward/record.url?scp=84893825073&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:84893825073

SN - 9781905088386

VL - 93

BT - Proceedings of the 10th International Conference on Computational Structures Technology, CST 2010

PB - Civil-Comp Press

ER -

Lee TH, Cho SK. Kriging metamodel based statistical global optimization technique. In Proceedings of the 10th International Conference on Computational Structures Technology, CST 2010. Vol. 93. Civil-Comp Press. 2010