The Study Of An Artificial Fish Swarm Optimization Algorithm And Its Application

Artificial fish swarm algorithm is a new type of adaptive searching optimization algorithm, which is put forward by the domestic scholars. It has attracted the attention of many scholars with the advantages of strong robustness and good adaptive capacity and the use of flexible and fast convergence speed and so on. However, the traditional basic artificial fish swarm optimization algorithm is easy to fall into local extreme value and occur precocious phenomenon, when it searches for a optimization in the larger regional. And the optimization of algorithm rely on the selection of fish’s visual and step, when the vision and the step value is larger, the algorithm has faster convergence speed in the initial search, however, in the later algorithm optimization, it easy to occur the phenomenon, that skip over and volatility near the extreme point, affect the searching precision; when the vision and step size is small, the algorithm will be better in the later optimization accuracy, however, the algorithm convergence speed is slow, in the early running, increased the time cost of the overall optimization. So far, many scholars have devote to the improvement of the algorithm, but also achieved some results. In this paper, improved the vision and step of the basic artificial fish swarm algorithm, and combined with the niche technology, put forward a new improved artificial fish swarm optimization algorithm, referred to as an improved artificial fish swarm algorithm(simply called IAFSA).The main contents of this paper are as follows:(1) Firstly, The basic principles of the artificial fish swarm optimization algorithm is summarized. At the same time, the four basic behaviors of artificial fish swarm algorithm is detailed description. Then, several kinds of common intelligent optimization algorithm is briefed description to analysis the similarities and differences between different algorithms.(2) A new improved artificial fish swarm optimization algorithm is proposed in this paper. designing a new strategy to improve the artificial fish’s visual and step length, combing with best individual preservation strategy and sharing mechanism niche technology.(3) After disigned the new improved artificial fish swarm algorithm,put it into the field of function optimization to testify the performance of the new algorithm.