Improved sequential niche genetic algorithm for multimodal optimization
Xinjie Yu · Journal of Tsinghua University(Science and Technology) · 2001
Two improved strategies are presented for the standard sequential niche (SN) algorithm which faced difficulties when it is used to solve the sophisticated optimization problems. The first method is the new elitist method, which replaces the worst individuals with the recorded peaks between the selection and the crossover of the genetic algorithm. The second method mutates the individual near the recorded peaks with relatively high probability, without decreasing their fitness. Four new SNs are presented by combining the standard SN with these two strategies. The massively deceptive problem is used to verify the strategies and algorithms. The results show that the methods effectively locate the total 32 global peaks.