The Improving Research On Wolf Search Algorithm With Ephemeral Memory

Optimization techniques have been widely applied in the domain of information, engineering and management presently. Generally, the optimization techniques usually include the modern swarm intelligence optimization algorithms and the traditional classical optimization methods that are based on the mathematical method. Swarm intelligence optimization algorithm has the advantage of overcoming the shortcomings which the traditional optimization methods usually have, that the adaptive scope is relatively narrow and nonscalable, and can be used to solve the complex and high-dimensional optimization problems. Now, Swarm intelligence optimization algorithm has become a frontier research topic in the domain of information. Wolf search algorithm with ephemeral memory(WSA) is a new proposed swarm intelligence optimization algorithm. Therefore, the research on improving the performance of WSA has both important theoretical value and practical application prospects. The work of this paper can be summarized as the following three parts:Firstly, aimed at the WSA that has weak global search ability and being easily fall into local optimum, this paper has proposed an improved WSA called an improved wolf search algorithm using Nelder-Mead operator. In the new improved algorithm, Nelder-Mead method is applied to the individual search activities. While searching for preys, every wolf can make full use of swarm information and individual memory to improve its search efficiency. What‘s more, those methods enhance the ability of the algorithm to avoid falling into local optimal, and improve the algorithm's global search ability and optimization performance.Secondly, based on the shortcomings of the wolf search algorithm with ephemeral memory such as: lack of the cooperating mechanism among individuals, use the group information not sufficient to guide its search activities, resist predators not with the power of collectives, an improved wolf search algorithm with ephemeral memory was proposed. Experiments were done on some typical benchmark functions, and the experimental results indicate that the proposed strategy improves the global search ability of the algorithm, and improves the optimization performance of the algorithm.Thirdly, in order to overcome the shortcomings of wolf search algorithm with ephemeral memory, a novel improved wolf search algorithm(IWSA) was proposed. And we applied it to solve the engineering optimization problems, such as data classification or clustering. The experimental results show that the improved algorithm has a better optimization performance when solving the data classification problem.