The Application of Advanced Operator Genetic Algorithms to Electromagnetic Optimization Problems

This dissertation discusses the application of genetic algorithms to problems of interest in electromagnetic engineering. Rather than simply apply a simple genetic algorithm to a given problem or overhaul genetic algorithms to make them more efficient for all problems, the approach taken in this thesis is to construct genetic algorithms that work well for certain classes of problems. Specifically, Pareto genetic algorithms are applied to the simultaneous optimization of several measures of merit of antenna arrays. Domain-decomposition genetic algorithms are introduced and applied to the optimization of E-plane microwave filters and frequency selective surfaces. Domain decomposition, the technique of breaking a given problem into parts and optimizing each part separately, is shown to accelerate both the analysis of the devices and the convergence of the genetic algorithm. To further accelerate the procedure, three novel model-order reduction techniques are introduced that enable efficient modeling of frequency selective surfaces and thus vastly enlarge the types of devices that can be designed in practice. Finally, dominance and diploidy structures are introduced into the genetic algorithm for application to the control of adaptive antenna arrays. All of the above-described design techniques are illustrated with practical examples, and the adaptive array control is demonstrated with simulations. Genetic algorithms are shown to be an efficient and effective technique for both off-line and on-line optimization of electromagnetic devices.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Electrical Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2015

Author dc:creator

• Weile, Daniel Seth

Contributors dc:contributor

• Eric Michielssen

• Engineering, Electronics and Electrical

Language dc:language

eng

Identifier

(MiAaPQ)AAI9945024

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/81299