An Improved Cuckoo Search Algorithm

Optimization algorithm is one of the theoretical tools to explore the optimal parameter values under some certain conditions.The maximum or minimum models can be used to characterize the practical problems.The optimization algorithm can solve these models appearing from many engineering fields,and satisfied results have been achieved.With the rapid development of social economy and equipment manufacturing industry,production cost minimization has become one of the key factors for enterprises to seize the market,win customers and obtain profits.Based on the cost savings,this paper has certain practical application value to research the Cuckoo Search algorithm in optimization problems.In this paper,there are mainly three aspects of research work as follows:(1)Study on the mathematical model of non-constrained and constrained optimization problems: many factors need to be considered in practical problems,and all kinds of factors have a certain mathematical relationship between them.The mathematical model of non-constrained optimization problem has the characteristics of progressive and continuous development.Therefore,based on the mathematical model of optimization problems in the research,numerical simulation has certain significance to the optimization problem.(2)The cuckoo search algorithm and its improved method research: This paper summarized the basic algorithm on the properties of the cuckoo algorithm,analyses the influence of the search mode and related parameters of cuckoo swarm algorithm convergence,and gives further a cuckoo search algorithm of the improved method.Mainly consists of the following three points.Firstly,the initial population is generated by the chaotic sequence,and the diversity of the population is increased;Secondly,the "teaching and learning" search method and cuckoo search algorithm Levy flight mode combined constructed new search method,effectively to balance the global search and local optimization;Finally,through the reflection method effectively solves the problem of the cross-border,the multi direction and multi angle searching of particles is realized.Two benchmark sets are tested to demonstrate that our improved cuckoo search algorithm has faster convergence speed and better stability.(3)This paper gives the mathematical model of improved cuckoo algorithm in non-constrained and constrained optimization problems on the example.By comparing with the basic cuckoo algorithm and other improved algorithms,it is shown that the improved algorithm in this paper has more advantages in the application of numerical optimization problem.Through the above research work,this paper argues that the hybrid cuckoo algorithm based on “teaching and learning” model performs better for solving numerical optimization problems.We think that the population initialization,improvement of search strategy and cross-border projection are the efficient ways to improve the cuckoo search algorithm.To a great extent,the convergence of the particle is accelerated,so it is better to solve the numerical optimization problems.