Fast optimization of electromagnetics design problems through the CMA evolutionary strategy

M. D. Gregory, Z. Bayraktar, D. H. Werner

Research output: Chapter in Book/Report/Conference proceedingConference contribution

3 Scopus citations

Abstract

The design of antennas, frequency selective surfaces, arrays, and other electromagnetic devices often requires tuning of many geometric and material parameters to obtain a desired set of characteristics. Typically, a genetic algorithm, particle swarm, or other standard optimization technique is used to obtain a suitable set of parameters that fulfill the design criteria. Although these methods have been shown to be relatively robust and reliable, they often require long optimization times or function evaluations, usually in the form of many simulations, to find an acceptable solution. Many new and improved evolutionary strategies have arisen since the inception of the genetic algorithm [1] and particle swarm techniques. The covariance matrix adaptation evolutionary strategy (CMA-ES) is a recently designed algorithm that has generated a great deal of interest in the evolutionary computation community. In addition to being a fast strategy, CMA also requires very few algorithm parameters due to its adaptive nature, eliminating issues with user setting selections [2]. In this paper, this new algorithm is applied to an antenna optimization problem and comparisons are made to a particle swarm technique [3], demonstrating the appealing properties of CMA that make it ideally suited for electromagnetics design.

Original languageEnglish (US)
Title of host publication2010 IEEE International Symposium on Antennas and Propagation and CNC-USNC/URSI Radio Science Meeting - Leading the Wave, AP-S/URSI 2010
DOIs
StatePublished - Nov 22 2010
Event2010 IEEE International Symposium on Antennas and Propagation and CNC-USNC/URSI Radio Science Meeting - Leading the Wave, AP-S/URSI 2010 - Toronto, ON, Canada
Duration: Jul 11 2010Jul 17 2010

Publication series

Name2010 IEEE International Symposium on Antennas and Propagation and CNC-USNC/URSI Radio Science Meeting - Leading the Wave, AP-S/URSI 2010

Other

Other2010 IEEE International Symposium on Antennas and Propagation and CNC-USNC/URSI Radio Science Meeting - Leading the Wave, AP-S/URSI 2010
CountryCanada
CityToronto, ON
Period7/11/107/17/10

All Science Journal Classification (ASJC) codes

  • Computer Networks and Communications
  • Hardware and Architecture

Fingerprint Dive into the research topics of 'Fast optimization of electromagnetics design problems through the CMA evolutionary strategy'. Together they form a unique fingerprint.

  • Cite this

    Gregory, M. D., Bayraktar, Z., & Werner, D. H. (2010). Fast optimization of electromagnetics design problems through the CMA evolutionary strategy. In 2010 IEEE International Symposium on Antennas and Propagation and CNC-USNC/URSI Radio Science Meeting - Leading the Wave, AP-S/URSI 2010 [5562217] (2010 IEEE International Symposium on Antennas and Propagation and CNC-USNC/URSI Radio Science Meeting - Leading the Wave, AP-S/URSI 2010). https://doi.org/10.1109/APS.2010.5562217