Title :
Electromagnetic design optimization using mixed-parameter and multiobjective CMA-ES
Author :
BouDaher, Elie ; Hoorfar, Ahmad
Author_Institution :
Dept. of Electr. & Comput. Eng., Villanova Univ., Villanova, PA, USA
Abstract :
In this paper, new variations of Covariance Matrix Adaptation Evolution Strategy (CMA-ES) are investigated for optimization of electromagnetic design problems. Two variations of the CMA-ES algorithm are presented: mixed-parameter CMA-ES and multiobjective CMA-ES. Examples and results are provided to evaluate the performance of these algorithms.
Keywords :
covariance matrices; electromagnetic wave propagation; genetic algorithms; covariance matrix adaptation evolution strategy; electromagnetic design optimization; mixed-parameter; multiobjective CMA-ES; Algorithm design and analysis; Covariance matrices; Electromagnetics; Evolutionary computation; Genetic algorithms; Linear programming; Optimization;
Conference_Titel :
Antennas and Propagation Society International Symposium (APSURSI), 2013 IEEE
Conference_Location :
Orlando, FL
Print_ISBN :
978-1-4673-5315-1
DOI :
10.1109/APS.2013.6710864
