Rahmat Samii 1997 Genetic Algorithms In Engineering Electromagnetics英文电子书.pdf

Genetic Algorithms in Engineering Electromagnetics J. Michael Johnson and Yahya Rahmat-Samii Department of Electrical Engineering University of California, Los Angeles 405 Hilgard Ave. Los Angeles, CA 90095-1594 Tel: (310) 206-4801 (JMJ); (310) 206-3847 (YRS) Fax: (310) 206-8495 E-mail: johnson@ ;rahmat@ http: Keywords: Genetic algorithms, absorbing media, antenna arrays, differentiable andor discontinuous regions, and often utilizes array synthesis, microstrip antennas, natural modes, radar target approximations or models of the true electromagnetic phenomena identification to conserve computational resources. These characteristics sorely test the capabilities of many of the traditional optimization tech- niques, and often require hybridization of traditional optimization 1. Abstract methods, if these methods are to be applied at all. This paper presents a tutorial and overview of genetic algo- This paper focuses on a relatively new approach to optimiza- rithms for electromagnetic optimization. Genetic-algorithm (GA) tion, called the genetic algorithm (GA). Genetic algorithms are optimizers are robust, stochastic search methods modeled on the robust, stochastic-based search methods, which can handle the concepts of natural selection and evolution. The relationship common characteristics of electromagnetic optimization problems between traditional optimizatio