An Dynamic Adaptive Dissipative Particle Swarm Optimization with Mutation Operation

An adaptive dissipative particle swarm with mutation operation (ADPSO) is presented that combines the idea of the particle swarm optimization with concepts of mutation from evolutionary algorithm. In this paper, the problem and improved of the dissipative particle swarm optimization are analyzed deeply. The improvement ADPSO adopts Cauchy mutation operation to escape from the attraction of local minimum. In order to balance between global and local search, the adaptive inertia weight strategy is introduced. The simulation experiments demonstrate that ADPSO can not only effectively escape from local minimum, but also enhance the capability to search the global optimization in the later convergence phase.