Improved Artificial Fish-school Algorithm
Fan Yu-jun, Dongdong Wang, Mingming Sun · Chongqing Shifan Daxue xuebao. Ziran kexue ban · 2007
In this paper the Improved Artificial Fish-school Algorithm is proposed based on the study of the Artificial Fish-school Algorithm.The preying behavior is improved by introducing the strategy of keeping the best individual,and this method prevents the degenerating of the best individual in colony.The method of accelerating individual local searching is put forward,and it is used to improve the swarming behavior and fish's following behavior in order to make the global optimal value to be shown faster.Without influence on the final results,the searching area is reduced based on the definition and properties of bijection,so that global searching is accelerated.Several computer simulation results show that the Improved Artificial Fish-school Algorithm has some advantages such as higher precision of solution,higher efficiency of optimization,faster convergence rate, and better stabilization etc.