Research And Application Of Artificial Fish Swarm Intelligent Optimization Algorithm

Traditional optimization algorithms often have some particular requirements for optimization problems' analytic property.In general,they are very hard to solve different kinds of complex and diverse optimization problems.In recent years,the study found that swarm intelligent optimization algorithms have unique advantages in sloving these optimization problems,it has been paid attention by many scholars.To be an intelligent algorithm of imitating the fish prey,Artificial Fish Swarm Algorithm(AFSA)has theses characteristics:objective function doesn't have any particular requirements;AFSA is insensitive to initial value and parameter values,and it has parallel processing ability and good randomness,etc.AFSA has been widly researched and applied in science and practicial engineering problems.In this paper,AFSA has been improved and applied in solving the problem of the logistics distribution center location problem.Specific content summarized as follows:(1)AFSA has its drawbacks such as falling into local optimum and low search accuracy.To solve these problems,this paper proposed an opposite adaptive and gauss mutation artificial fish swarm algorithm(OAGMAFSA).To provide more opportunities to explore potential better area,the algorithm applied opposite point to adjust direction and location of artificial fish.Thereby,the artificial fish can jump out of local optimum fast.In addition,this algorithm balanced the global and local searching ability by using a non-linear function to adjust artificial fish's visual and step.Moreover,in order to solve early-maturing of artificial fish,using gauss mutation mechanism based on optimal solution increased the diversity of every artificial fish.The simulation results show that OAGMAFSA has better accuracy.Meanwhile,the algorithm avoids early-maturing compared with other AFSAs.(2)The study found that there are two deficient aspects of the basic Arctificial Fish Swarm algorithm.First,it can't adjust visual and step adaptively according to fish's distribution in their preying process.Secondly,every artificial fish's behavior belongs to local searching,lacking of global superiority.To solve these problems,this paper proposed an elite learning-based Multi-dimensional dynamic adaptive artificial fish swarm algorithm(EMAAFSA).The improved algorithm set independent visual and step for each dimension,and defined visual vector,step matrix and multi-dimensional neighborhood,then improved 4 basic behaviors.Thereby,the artificial fish can adjust their own searching range adaptively according to their distribution.At the same time,in order to improve fish's global search performance and decrease the probability of falling into local optimum,this paper proposed an elite learning strategy.The simulation results show that EMAAFSA has good searching quality and and robustness.Meanwhile,EMAAFSA improves artificial fish's global searching ability compared with other AFSAs.(3)The logistics distribution center is in charge of storing lots of goods.The center is used to distribute and assemble goods.The multi logistics distribution center location problem is a constrained nonlinear programming problem.In this paper,improved artificial fish swarm optimization algorithms are applied to multi logistics distribution center location problem.The simulation results show that OAGMAFSA and EMAAFSA have good availability and application value.