Research On Nonlinear Swarm Intelligence Optimization And Application

In the research of scientific and the application of engineering technology,a large number of problems can be converted into optimization problems,which usually have the characteristics of complexity,non-linearity and non-differentiability.Traditional numerical optimization methods are inefficient to solve these problems.Research on effective optimization methods is conducive to the development of various technology and science.Swarm intelligence optimization algorithm is an important branch of intelligent optimization algorithm.It has the advantages of simple structure,fast convergence and easy implementation.However,swarm intelligence optimization algorithm has some limitations,especially for large-scale optimization problems,such as low accuracy,easy to fall into local optimum and other defects.Therefore,it is of great significance to design swarm intelligence optimization algorithm with better performance to cope with the problems of science and technology.Firstly,two classical swarm intelligence optimization algorithms are introduced concisely.Then,in each chapter,we focus on improving performance of particle swarm optimization algorithm and differential evolution algorithm to solve practical optimization problems.(1)A new multi-stage perturbed differential evolution(MPDE)is proposed in this paper.A new mutation strategy "multi-stage perturbation" is implemented with directivity difference information strategy and multiple parameters adaption.The DE/current-to-pbest is introduced to increase the population diversity while remaining its elitist learning behavior in this architecture.The multi-stage perturbation-based mutation operation utilizes the Normal random distribution with adjustable variance to perturb the chosen solutions.Multiple parameters are adaptively adjusted to appropriate values to match the current search status of algorithm.It is thus helpful to enhance the performance and the robustness of algorithm.Simulation results show that the newly proposed MPDE is better than,or at least comparable to other algorithms in terms of optimization performance based on benchmark function.(2)Differential evolution(DE)has attracted more and more attention.However,the neighborhood and direction information has not been fully utilized in exploration and exploitation stages.A failure remember-driven self-adaptive differential evolution algorithm,ATBDE,is proposed in this paper,which uses "Top-Bottom" strategy with optional archive and a parameter self-adapting strategy driven by "Failure Remember" operation."Top-Bottom"strategy utilizes historical heuristic information obtained from the successful and failed individuals,respectively,to guide individuals toward the potential more promising regions in an optional archive manner.This strategy is also theoretically analyzed for the best implementation.The failure remember-driven parameter adaption strategy shares the positive search experience from the successful individuals and abandons the negative search experience for those successive failing individuals.Experimental results show that ATBDE is better than,or at least comparable to other DE algorithms in terms of convergence performance and accuracy.(3)Particle swarm optimization(PSO)is a well-known evolutionary algorithm for its simplicity and effectiveness,which usually has strong global search capability and drawback of being easily trapped by local optima.A scaling mutation strategy and an elitist learning strategy are presented in this paper.Based on them,an improved PSO variant(LSERPSO)is accordingly proposed with local search and ring topology strategy.The new scaling mutation strategy is an exploration and exploitation balance focusing mutation operation.A collection of elitist individuals is maintained such that the following particles can learn from them.A ring topology based neighborhood structure is adopted to maintain the population diversity and to reduce the possibility of being trapped in local optima.Finally,Quasi-Newton based local search is incorporated to enhance the fine-grained capability.The effects of these proposed strategies and their cooperation are verified step by step.The performance of LSERPSO is comprehensively studied on benchmark functions.(4)This paper attempts to address the problem of large scale optimization and high dimensional optimization using principal component analysis(PCA)strategy with differential evolution(DE)based on Cooperative Co-evolution(CC)framework.The decomposition method is a maj or obstacle for large-scale optimization problems.The aim of this paper is to propose effective dimension decomposition method of PCA strategy for capturing the main information among dimensions.PCA strategy can measures most of the contribution information of dimension and uses it for identifying main dimension to guide them to group the most promising subcomponents in CC framework.Then each subcomponents can be solved using an evolutionary optimizer to find the optimum values.The experimental results show that this new technique is more effective than some existing grouping methods.(5)An improved particle swarm optimization algorithm is proposed to solve the job shop scheduling problem optimization.By analyzing the characteristics of job shop scheduling problem and according to various resource constraints,the operation based representation method and activity scheduling decoding method is designed for particle swarm optimization(PSO).Particle swarm optimization algorithm is improved by using a scaling mutation strategy and ring topology structure,and an elite selection learning strategy is added to retain excellent individual information and help accelerating convergence speed.This improved particle swarm optimization(PSO)algorithm is used to solve standard case problems of different scales and is compared with other methods to verify the effectiveness of the algorithm.