The Improved Method Research Of Particle Swarm Optimization

Based on the thorough analysis of the PSO algorithm deficiency and its cause, PSO is combined with cloud theory, Artificial Fish School algorithm in this paper. A new PSO based on expanded mutation is proposed to enhance the global exploration abilities and solve the premature convergence caused by some particles in standard PSO fall into stagnation.The main tasks of this article are as follows:(1)A cloud adaptive Particle Swarm Optimization algorithm based on the expanded mutation is proposed, in this algorithm, inertia weight of every particle adjusts based on the cloud model X-generator adaptively; The expanded mutation is used to overcome the difficulties cause by the multi-dimensional and multi-variable interference. It is an effective method for high-dimensional space problems. The simulation results show the algorithm is of high robust, fast convergence and high accuracy.(2)The particles are divided into 3 equal groups, The formulation of how particle adjusts its position adopt different rules based on Particle Swarm Optimization, A modified particle swam optimization and A Cloud Adaptive Particle Swarm Optimization respectively. The aim is to keep the independence of the particle swarm and the superiority of the optimization, but not to increase the complexity of the algorithm. Moreover, the super-society parts in which the velocity formula is redefined is put forward, and the expanded mutation method, disturbance Operation are proposed. The simulation results show that the new algorithm is more precise and stable, gets better results in a faster and cheaper way.(3) Based on artificial fish swarm algorithm (AFSA) and Particle Swarm Optimization (PSO), a hybrid particle swarm optimization algorithm is proposed. The method makes full use of the global convergent performance of AFSA and the quickly local convergent performance of PSO, and the method is used to optimize complex nonlinear function and solves complex chemistry equation roots. Finally, the numerical experiment results show that hybrid particle swarm optimization algorithm owns a faster convergence rate with a good global convergence performance.