Pareto-Ranking Bisection Algorithm for Expedited Multiobjective Optimization of Antenna Structures

The purpose of this letter is the introduction of a novel methodology for expedited multiobjective design of antenna structures. The key component of the presented approach is fast identification of the initial representation of the Pareto front (i.e., a set of design representing the best possible...

Full description

Saved in:
Bibliographic Details
Published in:IEEE antennas and wireless propagation letters Vol. 16; pp. 1488 - 1491
Main Authors: Koziel, Slawomir, Bekasiewicz, Adrian
Format: Journal Article
Language:English
Published: IEEE 2017
Subjects:
Algorithm design and analysis
Antenna design
Approximation algorithms
bisection algorithm
Computational modeling
multiobjective optimization
Optimization
Partitioning algorithms
response correction
simulation-driven design
surrogate modeling
Ultra wideband antennas
variable-fidelity simulations
ISSN:1536-1225, 1548-5757
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:The purpose of this letter is the introduction of a novel methodology for expedited multiobjective design of antenna structures. The key component of the presented approach is fast identification of the initial representation of the Pareto front (i.e., a set of design representing the best possible tradeoffs between conflicting objectives) using a Pareto-ranking bisection algorithm. The algorithm finds a discrete set of Pareto-optimal designs. Its operation principle is sequential partitioning of the line segments connecting the designs found in the previous iterations, and refining the new designs allocated this way by means of poll-type search involving Pareto ranking. Subsequently, the final Pareto set is obtained by means of response correction techniques. Our methodology is demonstrated using an ultrawideband monopole antenna and compared to state-of-the-art multiobjective optimization methods.
ISSN:1536-1225
1548-5757
DOI:10.1109/LAWP.2016.2646842