Abstract
In general, as diffractive optical elements formed by use of self-repeating patterns possess beneficial characteristics such as scratch resistance, low design effort, ease of fabrication, and natural formation of large panels, an efficient design methodology that was developed with a modified preserving-the-best strategy of genetic algorithms is presented. Both genetic algorithms and simulated annealing are examined by the Markov-chain stochastic process to create the insight needed to use these two heuristic algorithms efficiently. It was found that adding the preserving-the-best strategy to traditional genetic algorithms guarantees the possibility of locating the global optimum. Combining this sufficient and necessary condition for locating a global optimum for genetic algorithms with the built-in chromosome crossover searching mechanism and its neighborhood identification makes this newly developed genetic algorithm an effective method for designing diffractive optical elements. In our study, a prototype was fabricated based on our case study with the modified genetic algorithm. The performance of this prototype was measured and analyzed. Experimental results are shown to agree well with theoretical predictions.
© 1997 Optical Society of America
Full Article | PDF ArticleMore Like This
Guangya Zhou, Yixin Chen, Zongguang Wang, and Hongwei Song
Appl. Opt. 38(20) 4281-4290 (1999)
Aleksandra B. Djurišić, Jovan M. Elazar, and Aleksandar D. Rakić
Appl. Opt. 36(28) 7097-7103 (1997)
Chih-hao Chen and A. A. Sawchuk
Appl. Opt. 36(29) 7297-7306 (1997)