The Study Of Particle Swarm Optimization And Improvement

Particle Swarm Optimization algorithm, which is a kind of swarm intelligent optimization algorithms, is put forward by foraging act of research and observation about groups of birds. Because Its favorable characteristics, it's focused by the majority of scholars. In recent years, it becomes an important part of modern optimized method, and shows great potential in the target function ,optimization engineering practice and so on. In itself, there are still a lot of defect in theory and practice. In the light of the shortcomings of particle swarm optimization algorithm is easy to fall into local optimum, an improved particle swarm optimization and a hybrid optimization algorithm are obtained in this paper.First, particle swarm optimization based on period evolutionary tactics In new algorithm, When the algorithm may fall into a local optimum by judgments, the values of some particles are re-assigned by a certain proportion, so the algorithm is out of local optimum, near to the global extreme. Numerical experiments prove that the calculation of accuracy and convergence speed of the improved algorithm have been greatly improved.Second, the hybrid particle swarm optimization algorithm of particle and fish swarm The algorithm analyzes the respectively features of particle and fish swarm: particle swarm algorithm has fast convergence and can reserve optimum condition of every particle. As a result of introducing congestion factor, fish swarm algorithm has the ability of global optimization. Combining the advantages of them, hybrid optimization algorithm based on particle and fish swarm is made. It has satisfactory optimized performance through the test of the classical function.