Multi-fidelity optimization via surrogate modelling
Multi-fidelity optimization via surrogate modelling
This paper demonstrates the application of correlated Gaussian process based
approximations to optimization where multiple levels of analysis are available, using an
extension to the geostatistical method of co-kriging. An exchange algorithm is used to
choose which points of the search space to sample within each level of analysis. The
derivation of the co-kriging equations is presented in an intuitive manner, along with a new
variance estimator to account for varying degrees of computational ‘noise’ in the multiple
levels of analysis. A multi-fidelity wing optimization is used to demonstrate the
methodology.
co-kriging, kriging, noise, subset selection, wing design
3251-3269
Forrester, Alexander I.J.
176bf191-3fc2-46b4-80e0-9d9a0cd7a572
Sóbester, András
096857b0-cad6-45ae-9ae6-e66b8cc5d81b
Keane, Andy J.
26d7fa33-5415-4910-89d8-fb3620413def
8 December 2007
Forrester, Alexander I.J.
176bf191-3fc2-46b4-80e0-9d9a0cd7a572
Sóbester, András
096857b0-cad6-45ae-9ae6-e66b8cc5d81b
Keane, Andy J.
26d7fa33-5415-4910-89d8-fb3620413def
Forrester, Alexander I.J., Sóbester, András and Keane, Andy J.
(2007)
Multi-fidelity optimization via surrogate modelling.
Proceedings of the Royal Society A, 463 (2088), .
(doi:10.1098/rspa.2007.1900).
Abstract
This paper demonstrates the application of correlated Gaussian process based
approximations to optimization where multiple levels of analysis are available, using an
extension to the geostatistical method of co-kriging. An exchange algorithm is used to
choose which points of the search space to sample within each level of analysis. The
derivation of the co-kriging equations is presented in an intuitive manner, along with a new
variance estimator to account for varying degrees of computational ‘noise’ in the multiple
levels of analysis. A multi-fidelity wing optimization is used to demonstrate the
methodology.
Text
RSPA20071900.pdf
- Version of Record
More information
Published date: 8 December 2007
Keywords:
co-kriging, kriging, noise, subset selection, wing design
Identifiers
Local EPrints ID: 64698
URI: http://eprints.soton.ac.uk/id/eprint/64698
ISSN: 1364-5021
PURE UUID: f5bb57a4-d2f0-4d1a-8283-d75456b5c831
Catalogue record
Date deposited: 09 Jan 2009
Last modified: 16 Mar 2024 03:26
Export record
Altmetrics
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics
