Font Size: a A A

Improved Research And Application Of Moth-flame Optimization Algorithm

Posted on:2021-08-19Degree:MasterType:Thesis
Country:ChinaCandidate:Q LiuFull Text:PDF
GTID:2518306458992759Subject:Computer application technology
Abstract/Summary:PDF Full Text Request
The Moth-flame optimization algorithm(MFO)is a novel swarm intelligence algorithm proposed by Seyedali Mirjalili in 2015.Moths and flames are the two components of the algorithm,and the balance between exploration and exploitation can be solved by the horizontal positioning and navigation mechanism.The Moth-flame optimization algorithm has the advantages of simple parameters and easy implementation.Therefore,since MFO was proposed,it has become one of the research hotspots and has been applied in many fields.However,the MFO algorithm has shortcomings such as premature convergence,easy to fall into local optimization and low optimization accuracy.To some extent,these shortcomings limit its application.In this paper,some improvements have been made and the improved algorithm is used to solve the real-world optimization problem and K-means clustering problem.Its purpose is to improve the performance of MFO algorithm and expand its application scope.Specifically,the main achievements of this paper include the following two aspects:1.Aiming at the premature convergence and low accuracy of MFO algorithm,an improved MFO algorithm based on chaotic initialization and Gaussian mutation is proposed(CGMFO).First,the chaotic cube map is used to initialize the moth population to make the moths more evenly distributed in the search space.Secondly,Gaussian mutation is used to perturb a small number of poor individuals in the population to enhance the ability to jump out of the local optimal.Lastly Archimedes curve is used to expand the search range and is able to explore unknown areas.A series of experimental results show that the improved algorithm can significantly improve the accuracy of the solution and the convergence speed of the algorithm.Additionally,the improved MFO algorithm is applied to the real-world engineering optimization problem and has achieved good results.2.The traditional K-means algorithm is very sensitive to the selection of initial clustering centers and is easy to fall into local optimization.Therefore a dynamic adaptive moth flame optimization algorithm with multi-strategies is proposed(MMFO).The algorithm integrates opposition-based learning strategy,elite strategy and dynamic adaptive weighting strategy in the MFO algorithm to enhance the global search ability,convergence speed and solution accuracy.The experimental results demonstrate that the reliability and stability of the clustering can significantly be improved when applying the improved MFO algorithm.
Keywords/Search Tags:Moth-flame optimization algorithm, Chaos mapping, Gaussian mutation, Opposition-based learning, Elite strategy, K-means algorithm, Engineering optimization
PDF Full Text Request
Related items