Skip to Main Content
With many advantages of computing with real number, few parameters to be adjusted, the Particle Swarm Optimizer (PSO) is applied in many fields. The major problem with PSO algorithm is premature convergence. Some optimization strategies were introduced to overcome it. In these former researches, the dimension of benchmarks in experiments was usually set to be a small value. But it can be seen that when the benchmark is with high dimension, the basic PSO and some advantage versions can not converge to a satisfied point. This paper presents a new particle swarm optimizer algorithm-AF-PSO. The AF-PSO uses the adaptive-trying strategy to accelerate the particle swarm convergence speed. To avoid premature convergence of the swarm, adaptive-mutation is also adopted. The HPSO is compared with the BPSO and GCPSO, the experiment result shows that the new algorithm performances better on a four-function test suite with high-dimension.