A study of genetic algorithms based on isolated niche technique
Teng Hai-wen · He'nan kexue · 2005
Simple genetic algorithms get local optimal solution too easily and converge slowly.To solve these problems,isolation mechanism and self adapting technique are studied in this article,and we find isolation mechanism plays important roles in evolution and can actively improve the diversity of populations,while crossover rate and mutation rate are not only related to fitness function value but also affected greatly by evolution eras.In this article,a new adaptive genetic algorithms based on isolated niche technique is presented,and classical and practical examples simulated.The results show that the adaptive genetic algorithms based on niche technique has greatly improve the diversity of populations,quicken the speed of convergence and enhance the probability of getting the best solution.