A new sample update strategy based on kringing

Yuanyuan Chen, Junqiang Bai, Dan Wang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

The study presented in this paper sets up a new mixing sample update strategy based on the Kriging model to address the issue of accuracy in complicated engineering optimization problems by adding points dynamically. First, a new optimized Latin Hypercube Sampling experiment design method is established to increase the homogeneity of the sample. Second, a sample update strategy based on local optimization and EI-based Kriging model is established. Three numerical samples are carried out to compare the results of traditional popular method EGO, and the results show that this method has increased the accuracy and the efficiency. The successful application of this method on RAE2822 Airfoil using less sample and reducing 14.3% of the drag has further proven the effectiveness of this method.

Original languageEnglish
Title of host publication26th Chinese Control and Decision Conference, CCDC 2014
PublisherIEEE Computer Society
Pages1124-1131
Number of pages8
ISBN (Print)9781479937066
DOIs
StatePublished - 2014
Event26th Chinese Control and Decision Conference, CCDC 2014 - Changsha, China
Duration: 31 May 20142 Jun 2014

Publication series

Name26th Chinese Control and Decision Conference, CCDC 2014

Conference

Conference26th Chinese Control and Decision Conference, CCDC 2014
Country/TerritoryChina
CityChangsha
Period31/05/142/06/14

Keywords

  • EGO
  • Kriging
  • Latin Hypercube Sampling
  • Optimization

Fingerprint

Dive into the research topics of 'A new sample update strategy based on kringing'. Together they form a unique fingerprint.

Cite this