Improvement And Application Of Hybrid Particle Swarm Optimization

In engineering science,There are a lot of practical applications are applied to solve the problem can be transformed into optimization problem.The Particle Swarm Optimization(PSO)algorithm does not need to establish a detailed mathematical model,but rather through simulating the flock foraging behavior of a kind of intelligent algorithm,to the realization of the optimal solution of complex problems.It has a simple structure,few parameters,fast convergence speed and is easy to implement.However,particle swarm optimization is easy to fall into premature convergence and can' t find the optimal solution.There is a shortage of poor population diversity in the later stage of convergence.Therefore,it is of great practical significance to further study the improvement of PSO and its application.The main research work of this paper is as follows:First,PSO the optimization process,parameter settings,and several common improved PSO are introduced,and the convergence principle of the particle swarm is analyzed.In view of easy to premature convergence of PSO,Combine with the Beetle Antennae Search Algorithm(BAS)algorithm and the Simple Particle Swarm Optimization(SPSO)algorithm,the step size of particles trapped in local extremum is jumped,and the jumping particles are used as a new information source to make other particles learn again.An improved PSO algorithm based on the step of the beetle is proposed to improve the algorithm accuracy.Secondly,The algorithm uses chaos to initialize the population,combines with the difference grouping in the Shuffled Frog Leading Algorithm,based on the fact that quantum particle swarm can search globally.A chaotic quantum particle swarm optimization algorithm based on leapfrog strategy is proposed.The algorithm enhances the information interaction between particles,and improves the situation that the population diversity is poor and prone to local extremum in the later stage of the search.In order to increase the convergence rate of particle swarm optimization,a chaotic PSO algorithm based on leapfrog strategy is proposed.The test function is used to simulate the improved particle swarm by MATLAB simulation software,and it is proved that the improved algorithm can further improve the search accuracy and search performance of algorithm.Finally,the improved PSO was used to optimize the maximum inter-class variance method to establish a multi-threshold segmentation model for images.Four test images were selected for simulation test,and the optimal algorithm model was obtained through data comparison,so as to optimize the multi-threshold segmentation of images.Then,an improved PSO neural network model is established to optimize the initial weights and thresholds of the neural network,so as to increase the performance of the network.The convergence performance and optimization speed of the PSO algorithm are verified by applying the optimized model to weather prediction...