Genetic algorithms as applied to the numerical computation of electromagnetic scattering by weakly nonlinear dielectric cylinders

This paper deals with the application of an optimization procedure based on a genetic algorithm (GA) to the prediction of the electromagnetic fields scattered by weakly nonlinear dielectric objects. Starting by an integral approach and describing the nonlinearities of the constitutive parameters by the Volterra-type integrals, the nonlinear scattering problem is numerically solved by an iterative procedure developed for the minimization of a suitable defined cost function. A GA is applied in order to deal with a large number of unknowns related to the harmonic components of the nonlinear internal electromagnetic field. In a preliminary stage, the behavior of typical parameters of the GA is analyzed; then numerical solutions are carried out and compared with those provided by other methods. Finally, some considerations are made concerning the rate of convergence of the iterative procedure.