Research Of Parallel Particle Swarm Optimization Based On Multi-population

As a new type of swarm intelligent, Swarm Intelligence Algorithms has attracted more and more researchers’attention. It simulates the swarm intelligence behavior of gregarious displayed by mutual cooperation, which contains features such as distribution, self-organization, cooperation, robustness and complement. Meanwhile, it also provides reliable and convenient resolutions for basic theoretical problems.Although the PSO algorithm has good effect for simple optimization problems, in the optimization of complex high-dimensional problems with multiple extreme points, the algorithm is prone to "premature" convergence in local optima. Therefore, how to avoid particles falling into local optima is an important research subject of the algorithm.This paper based on particle swarm optimization algorithm analysis and research, found that the lack of standard particle swarm optimization algorithm. According to the different characteristics of particle swarm evolutionary population, based on multi-population parallel particle swarm optimization algorithm, to increase the population diversity, improve the algorithm global convergence. Finally, the improved algorithm is applied to deployment of mobile nodes. This article contains work as follows:(1)Analysis of Particle Swarm Optimization algorithm populationWhen the particle swarm optimization algorithm in some parameters obtained from different values, obtained different evolutionary model, the evolution model of algorithm to produce different convergence effect.(2) Research of Parallel Particle Swarm Optimization based on multi-populationBecause the standard particle swarm optimization algorithm tends to "premature" convergence, the particle groups is divided into a number of clustering, different groups using different populations of parallel evolution. It can make full use of the evolutionary population characteristics, so the particles in the process of evolution have different flight characteristics, so as not to cause the diversity of population decreased rapidly to fall into the local extreme value. (3)Improved PSO algorithm applied to wireless sensor nodes deploymentFinally, in the view of the traditional deployment methods of mobile nodes in WSN, there are some problems like non-uniform distribution and incomplete coverage of the nodes. To overcome this, a new assignment deployment method of mobile nodes, which is based on Particle Swarm Optimization Algorithm (PSO), is presented from the angle of swarm intelligence algorithm. Simulation results show that the improved PSO has stronger global convergence ability than standard PSO in the application of deployment of mobile nodes, and reduce the wasted energy of the mobile nodes at the same time.