Interval multi-objective optimisation of structures using adaptive Kriging approximations

► An interval uncertain multi-objective optimisation (IUMOO) method is proposed. ► Interval model is used to describe structures with uncertain-but-bounded parameters. ► An adaptive approximation scheme is proposed using the Kriging model and LHD method. ► The adaptive model is sequentially updated...

Full description

Saved in:
Bibliographic Details
Published in:Computers & structures 2013-04, Vol.119, p.68-84
Main Authors: Li, Fangyi, Luo, Zhen, Rong, Jianhua, Zhang, Nong
Format: Article
Language:English
Subjects:
Approximation
Approximation model
Calibration
Computational efficiency
Interval number programming
Intervals
Kriging
Mathematical models
Multi-objective optimisation
Optimization
Sampling
Uncertainty
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:► An interval uncertain multi-objective optimisation (IUMOO) method is proposed. ► Interval model is used to describe structures with uncertain-but-bounded parameters. ► An adaptive approximation scheme is proposed using the Kriging model and LHD method. ► The adaptive model is sequentially updated via the space-filling design criteria. ► Interval number programming method is used to generate deterministic optimisations. This paper proposes an interval uncertain multi-objective optimisation (IUMOO) method for structures with uncertain-but-bounded parameters. An adaptive Kriging model is established to improve the computational efficiency and numerical accuracy in the approximation of design functions. Latin Hypercube Design (LHD) is applied to achieve a set of sampling points both in the design and uncertain spaces for calibrating the Kriging surrogate model. The interval number programming method is used to transform the uncertain optimisation into a corresponding deterministic multi-objective optimisation. Typical numerical examples are used to demonstrate the effectiveness of the proposed methodology.
ISSN:0045-7949
1879-2243
DOI:10.1016/j.compstruc.2012.12.028