Simulation-Driven Modeling and Optimization by Slawomir Koziel Leifur Leifsson & Xin-She Yang

Author:Slawomir Koziel, Leifur Leifsson & Xin-She Yang
Language: eng
Format: epub
Publisher: Springer International Publishing, Cham

3.2 Surrogate Models for Narrowband Antenna Design

Considerable numerical cost of evaluating the high-fidelity model R f makes its direct multi-objective optimization prohibitive. Computational speedup can be obtained using surrogate-assisted techniques [1, 7, 33], where the optimization burden is shifted into the cheaper low-fidelity representation of the structure of interest referred to as R c . For the sake of brevity we omit here the detailed description of the low-fidelity model setup. The details can be found in [23, 29].

Evaluation cost of the considered DRA antenna model (cf. Section 2) is reduced by means of coarse mesh discretization. Consequently, the R c is lacking accuracy and it has to be corrected to become a reliable representation of R f , especially in terms of its return loss response (other characteristics, e.g., gain, are normally well aligned for both R f and R c ). In case of narrowband antennas such as the considered DRA, the major type of discrepancy is frequency shift which can be reduced by suitable frequency scaling.

The frequency-scaled R c model (denoted as R cF ) is defined as

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.