Variable Velocity Limit Chaotic Particle Swarm Optimization

This paper analyzes the effects and the relationship of the parameters in the velocity updating of PSO. To velocity limit, the decreasing with different patterns are applied to play the varying part in the whole searching process. And the properties of the patterns are discussed. To overcome the shortage of the chaotic series, the logistic series and the improved tent series are introduced to replace the random numbers in PSO searching process. And the Variable Velocity Limit Chaotic Particle Swarm Optimization(VVL-CPSO) is proposed. In the numerical experiment, several benchmarks are applied to test the capabilities of the algorithms. The exploring ratio of the feasible region is defined. And the results show that the VVL-CPSO is better than the standard PSO and the Chaotic PSO. The VVL-CPSO, which is based on the improved tent series, has the best searching results on most of the benchmarks than other algorithms.