摘要
针对基本蛙跳算法搜索速度和精度不高的缺点,将变异的思想融入基本蛙跳算法,提出了一种新的改进蛙跳算法——交叉变异蛙跳算法(KSFLA).该算法通过子族群中排名较前的个体变异产生新个体代替子群中较差的若干个体,而其余的非子群最优个体则模仿大雁飞行策略,参照前面的个体进行位置更新,充分利用群体的信息提高算法性能.实验表明,改进后的算法在收敛速度以及收敛精度方面都有了很大程度地提高.
Aiming at the search speed and accuracy of the basic frog leaping algorithm not high,the idea of variation was integrated into the basic frog leaping algorithm.A new improved shuffled frog leaping algorithm was proposed which was called crossing and variation frog leaping algorithm.In the algorithm,the individuals ranked in front of the sub-populations were varietied to produce new individuals in order to instead of the poor individuals in the sub-populations.The location of other individuals not contained the best one in sub-populations was updated referring to the ones ranked in front of them like goose flying.This can make full use of the information of frog population groups.The experiments results reveal that the improved algorithm is better than the basic one in convergence velocity and convergence precision.
出处
《鲁东大学学报(自然科学版)》
2015年第1期16-20,共5页
Journal of Ludong University:Natural Science Edition
关键词
蛙跳算法
变异
收敛速度
收敛精度
SFLA
variation
convergence velocity
convergence precision