The paper presents a new particle swarm optimization algorithm with adaptive mutation operator. In the algorithm, a new adaptive mutation operator is given by fitness variance and space position aggregation degree and implemented at the best position of each particle at each iteration. The experiments on six problems show that the modified PSO algorithm can increases diversity of swarm and greatly enhance the ability to leap the local optimization.
Published in:
Information and Computing Science, 2009. ICIC '09. Second International Conference on
(Volume:1
)
Date of Conference: 21-22 May 2009