Font Size: a A A

Research On New Intelligent Optimization Algorithm And It's Application

Posted on:2020-08-06Degree:MasterType:Thesis
Country:ChinaCandidate:Z C SongFull Text:PDF
Abstract/Summary:PDF Full Text Request
The optimization problem is common in various fields of scientific research and engineering application.Therefore,the research on the optimization method has important theoretical significance and practical value,and it is still favored by scholars at home and abroad.A variety of algorithms have been developed with in-depth research on various optimization problems.The first is traditional algorithms such as linear programming and dynamic programming.Due to its top-down limitations,the development of traditional algorithms almost reaches the bottleneck.The emergence of random optimization algorithms gives this research a new vitality,and the group intelligent optimization algorithm is one of them.The so-called group intelligent optimization algorithm is a self-organizing and adaptive algorithm obtained by simulating the biological evolution process,which belongs to the category of artificial intelligence technology.The fruit fly optimization algorithm(FOA)and pollen algorithm(FPA)studied in this paper are new group intelligent optimization algorithms proposed in recent years.The former simulates the behavior of fruit fly group searching for food,requires fewer parameters,is extremely easy to understand,and is simple to implement.The latter simulates the pollination process and better balances the global optimization process with the local optimization process through novel mechanisms.Due to the shortcomings of FOA,such as easy to fall into local optimum and slow convergence in the later stage,this paper draws on the ideas of the improved algorithm,proposes a new improved method and applies it to the function optimization problem.Similarly,the FPA is also improved and applied to the problem of optimal load distribution of thermal power units.Through comparison experiments,they have achieved good results.The main research work and innovations of this paper are summarized as follows:(1)This paper introduces the basic principles of basic fruit fly optimization algorithm(FOA)and basic pollen algorithm(FPA),and gives the corresponding algorithm description and flow chart.Then it analyzes the advantages and disadvantages of each algorithm,making a theoretical basis for future research work(2)and(3).(2)Based on the introduction of FOA,this paper proposes a joint linear improved fruit fly optimization algorithm(LIFOA).Focusing on the new mechanism such as a linear generation mechanism,new odor concentration determination formula,weight system and synergy mechanism.The improved method improves the original algorithm by means of competition and cooperation,and can expand the optimization range of the algorithm,improve the solution precision and later convergence speed of the algorithm.Then the algorithm is applied to the function optimization problem.Then the algorithm is applied to the function optimization problem.Compared with FOA,LGMSFOA and SAMFOA,the comparison results with 16 benchmark functions show that the algorithm has high precision,fast convergence and strong global optimization ability when dealing with function optimization problems.(3)Based on the introduction of FPA,this paper proposes an improved pollen algorithm(IFPA),draws on the variation strategy of differential evolution algorithm(DE),introduces the cross-pollination process for the optimal individual and strengthens the local optimization ability of the algorithm.Then it is applied to the problem of optimal load distribution of thermal power units.The algorithm is used to solve the optimal distribution problem of the thermal power units of the 3,6,13 and 40 units of different total load requirements and the simulation experiment is carried out.The results show that the proposed algorithm has better performance and better value than pollen algorithm(FPA)and particle swarm optimization(PSO).
Keywords/Search Tags:fruit fly optimization algorithm, flower pollination algorithm, function optimization, thermal power unit load optimization distribution
PDF Full Text Request
Related items