Research On Improved Monarch Butterfly Optimization Algorithm And Its Application

The swarm intelligence optimization algorithm is a kind of bionic random method inspired by natural phenomena and biological behaviors and can deal with certain high-dimensional complex and variable optimization problems because of its better computing performance and simple model.Monarch Butterfly Optimization(MBO)algorithm is one of the swarm intelligence optimization algorithms,which simulates the migration behavior of monarch butterfly in nature.Since the MBO has few parameters and is easy to implement,it has been widely used in many fields,such as 0-1 knapsack problem,neural network training,dynamic vehicle routing problem,and classification of osteoporosis.In this paper,the research background and significance of MBO are introduced.The steps of MBO are described.The main defects of MBO are analyzed.The related studies of MBO are briefly reviewed.In order to further improve MBO and expand its application,aiming at the problems of MBO,two MBO's variants are proposed and applied to K-means clustering optimization problem.The main work of this paper is summarized as follows:(1)In order to solve the problems of MBO algorithm,such as it is easy to fall into local optimum and the convergence speed is low,this study proposed an improved MBO algorithm with cross migration and sharing adjustment(CSMBO).Firstly,this paper introduced a dimension-based vertical crossover operation to substitute the original migration operation of MBO,and then generated a cross migration operator.Thus,this operation could improve search ability of MBO algorithm effectively.Secondly,in order to speed up the convergence of MBO algorithm,the sharing adjustment operator with information sharing replaced the original adjustment operator.Finally,this paper utilized the greedy strategy to instead of the elite strategy of MBO,which could reduce one sorting operation and improve the calculation efficiency of MBO algorithm.To evaluate the optimization ability of our CSMBO algorithm,this paper made some experiments on a set of common benchmark functions with 30-dimensions and 50-dimensions,and the results showed that the proposed CSMBO algorithm had good optimization performance,and outperformed currently available three optimization approaches with which it is compared.(2)In order to solve the problems of MBO algorithm,such as poor performance on complex optimization problems,an improved MBO algorithm based on opposition-based learning and random local perturbation(OPOBO)is proposed.Firstly,the opposition-based learning method is introduced to generate the opposition-based population coming from the original population.By comparing the opposition-based population with the original population,the better individuals are selected and pass to the next generation,and then this process can efficiently prevent the MBO from falling into a local optimum.Secondly,a new random local perturbation is defined and introduced to improve the migration operator.This operation shares the information of excellent individuals and is helpful for guiding some poor individuals towards the optimal solution.The greedy strategy is employed to replace the elitist strategy to eliminate setting the elitist parameter in the basic MBO,and it can reduce a sorting operation and enhance the computational efficiency.A lot of experiments have been carried out on the benchmark functions of different dimensions,the results show that the OPMBO algorithm has higher optimization efficiency.(3)In order to further verify the performance of the improved MBO algorithm,the OPMBO algorithm is applied to deal with clustering optimization problem.The simulation experiments are carried out on 13 clustering datasets.The experimental results verify the universality of the OPMBO algorithm,which indicates that the OPMBO algorithm has good performance in dealing with clustering optimization problem.