A Global Artificial Fish Swarm Algorithm with Improved Chaotic Search

摘要

Artificial Fish Swarm Algorithm (AFSA) is an intelligence algorithm with advantages of insensitive to initial values, strong robustness and so on. However, it may be trapped in local optimum in the later evolution period and its search accuracy is dependent on step length which is hard to keep balance between rapidity and accuracy. A novel global artificial fish swarm algorithm is proposed in this paper, in which normal chaotic search on earlier stage is modified, and a differential evolution with improved chaos search was proposed to lead artificial fish into global optimum value.