An adaptive response surface method combined with (1+λ) evolution algorithm and its application to optimal design of electromagnetic devices

Author

Zhang, Yanli ; He, Houjian ; Koh, Chang Seop

Author_Institution

Sch. of Electr. Eng., Shenyang Univ. of Technol., Shenyang, China

fYear

2010

fDate

15-17 June 2010

Firstpage

2216

Lastpage

2220

Abstract

This paper presents a global optimization algorithm combining an adaptive response surface approximation of the objective function and (1+λ) evolution algorithm. In the adaptive approximation, an optimal Latin hypercube sampling strategy based on multi-objective Pareto optimization is developed to obtain the sampling data in the design variable space, and multiquadric radial basis function is employed to construct the response surface. The proposed method is tested and applied to the optimal design of permanent magnet linear synchronous motors.

Keywords

Pareto optimisation; design engineering; electromagnetic devices; response surface methodology; (1+λ) evolution algorithm; adaptive approximation; adaptive response surface approximation; adaptive response surface method; design variable space; electromagnetic devices; global optimization; multiobjective Pareto optimization; multiquadric radial basis function; objective function; optimal Latin hypercube sampling strategy; optimal design; permanent magnet linear synchronous motors; Algorithm design and analysis; Approximation algorithms; Electromagnetic devices; Finite element methods; Hypercubes; Interpolation; Response surface methodology; Sampling methods; Shape; Synchronous motors; Latin hybercube design; Pareto-optimization; central composite design; evolution algorithm; response surface method;

fLanguage

English

Publisher

ieee

Conference_Titel

Industrial Electronics and Applications (ICIEA), 2010 the 5th IEEE Conference on

Conference_Location

Taichung

Print_ISBN

978-1-4244-5045-9

Electronic_ISBN

978-1-4244-5046-6

Type

conf

DOI

10.1109/ICIEA.2010.5515906

Filename

5515906