Sequential inference method of electromagnetic optimization in engineering

被引：0

作者：

Li, Yanbin ^{[1
,2
]}

Guo, Yanqing ^{[2
]}

Lei, Gang ^{[1
]}

Shao, Keran ^{[1
]}

机构：

[1] College of Electrical and Electronic Engineering, Huazhong University of Science and Technology, Wuhan 430074, China

[2] School of Electronic Information, Zhongyuan University of Technology, Zhengzhou 450007, China

来源：

Huazhong Keji Daxue Xuebao (Ziran Kexue Ban)/Journal of Huazhong University of Science and Technology (Natural Science Edition) | 2009年 / 37卷 / 04期

关键词：

Approximate model - Computational effort - Electromagnetic optimization - IEEE TEAM Workshop Problem - Inference methods - Optimization algorithms - Radial basis - Team workshop problem 22;

D O I：

暂无

中图分类号：

学科分类号：

摘要：

Sequential inference method was presented for engineering electromagnetic optimization by the combination of statistical approximate models with sequential sampling method. In the implementation, approximate models include radial basis model and compactly supported radial basis model; sequential sampling and optimization process is composed of coarse optimization process and fine optimization process. Compared with traditional direct optimization method, the optimization models constructed by the sequential sampling method only need a small sample data, and the overall computational effort needed is much less than that by direct optimization method. Meanwhile, sequential method can provide prior information for the optimization algorithm. Finally, to illustrate the performance of the new method, Monte Carlo simulation experiment and the IEEE TEAM Workshop Problem 22 are investigated. Analysis demonstrates that the computation cost of the new methods is less than 1/10 of direct optimization algorithm. The optimization results are also satisfied the design objectives.

引用

页码：25 / 28