A Dynamical Particle Swarm Algorithm with Dimension Mutation

摘要

In this paper, a dynamical particle swarm algorithm with dimension mutation is proposed. First, we design a dynamically changing inertia weight which can change dynamically based on the speed factor and the accumulation factor. The algorithm with dynamically changing inertia weight can solve complex and nonlinear optimization process that linearly decreasing weight algorithm (LDW) is not adapt. Second, in order to escape from the local optimum, a dimension mutation operator is designed. The degrees of convergence of every dimension are calculated from the beginning of mutation. The dimension of the minimal convergence degree is mutated according to some probability. Finally, the simulation experiments also prove its high efficiency