Title :
A parametric study on differential evolution based on benchmark electromagnetic inverse scattering problem
Author_Institution :
Nat. Univ. of Singapore, Singapore
Abstract :
Electromagnetic inverse scattering problem is of great scientific and engineering importance. Differential evolution has been successfully applied to solve electromagnetic inverse scattering problem. However, it is noted that the performance of differential evolution strongly depends on its intrinsic control parameters, namely, population size, mutation intensity, and crossover probability. A parametric study on differential evolution based on benchmark electromagnetic inverse scattering problem is therefore carried out. The findings presented here should be highly beneficial to future applicants of differential evolution in the community of electromagnetic inverse scattering problems. Such findings may be directly or indirectly applicable to researchers in other areas.
Keywords :
electrical engineering computing; electromagnetic wave scattering; inverse problems; crossover probability; differential evolution; electromagnetic inverse scattering problem; mutation intensity; population size; Electromagnetic scattering; Evolutionary computation; Inverse problems; Parametric study;
Conference_Titel :
Evolutionary Computation, 2007. CEC 2007. IEEE Congress on
Conference_Location :
Singapore
Print_ISBN :
978-1-4244-1339-3
Electronic_ISBN :
978-1-4244-1340-9
DOI :
10.1109/CEC.2007.4424706