An Improved Artificial Fish Swarm Algorithm And Its Applications

Optimization problem has always been a hot research topic in the field of science and technology and the engineering application, the traditional optimization method in solving increasingly complex and scale increasing optimization problems encountered the inevitable problems. With the rapid development of computer technology, swarm intelligence algorithm as a novel and effective method in the field of optimization, many practical intelligent algorithm is put forward in succession, and the artificial fish swarm algorithm (AFSA) is one of the typical swarm intelligence algorithms.Through the research and analysis of AFSA’s principle and action mechanism, an improved algorithm was proposed based on the water flow mechanism which improves the search accuracy and global searching performance of the original AFSA, then applied it to combinatorial optimization problems, explored the new idea of AFSA’s application.In this paper, main work was as follows:(1) The basic concept and working principle of AFSA was introduced in detail, the influence of different parameters to AFSA’s performance was analyzed and the experience and guiding ideology during the initialization of parameters was summarized.(2) Did some research about the improvement scheme of AFSA, introduced some typical optimization strategy to AFSA and summed up the existing defects. Based on the research, an improved artificial fish swarm algorithm base on water flow mechanism was proposed. Through the simulation of nature water features-continuous flow and periodic flow, the ability of convergence accuracy and global convergence of AFSA was greatly improved.(3) Applied the improved AFSA to a typical combinatorial optimization problem-traveling salesman problem. Based on some targeted adjustments, we can use AFSA to solve TSP problem. Through the simulation experiment, the effectiveness of AFSA was proved and the application scope of AFSA was expanded.