I- Introduction

II- Central particle swarm optimization algorithm

III- Experiments

IV- Conclusions

REFERENCES

Abstract

The linear decreasing weight particle swarm optimization algorithm (LDWPSO) is mentioned in the concept of a center particle, and then puts forward center particle swarm optimization algorithm (PSO). The linear decreasing weight particle swarm optimization algorithm, unlike other general center particle, particle velocity center is not clear, and is always placed in the center of the particle swarm. In addition, the neural network training algorithm compared to particle swarm optimization algorithm and the linear decreasing weight particle swarm optimization algorithm, results show that: the performance is better than the linear optimization center particle swarm decreasing weight PSO algorithm. algorithm.

Introduction

PSO algorithm is a kind of method like social behavior and imitate birds and fish evolved from computing technology[1-3], PSO algorithm has good convergence and good performance in nonlinear function optimization. So it is getting more and more attention. Many researchers are committed to improving his performance with a variety of different methods and advanced interesting variables. These methods are broadly classified into the following categories. The improved method is to add a new coefficient to the velocity and position equation of the particle swarm optimization algorithm[4]. Finally, the coefficient should be selected reasonably. Angeline points out that the local search capability of the elementary particle swarm optimization algorithm is very low. To overcome this shortcoming, Shi and Eberhart proposed a LDWPSO algorithm[5], which introduced linear decrement inertial factor into the velocity update equation of the basic PSO algorithm.