Particle swarm optimization algorithm based on velocity differential mutation

To deal with the problem of premature local convergence, slow search speed and low convergence accuracy in the late evolutionary, this paper proposes a particle swarm optimization algorithm based on velocity differential mutation (VDMPSO). Firstly, The cause of local convergence in the basic PSO algorithm is elaborated. Secondly, strategies of direct mutation for the particle velocity rather than the traditional particle position with differential evolution algorithm based on analyzing the relations of the particle velocity and the population diversity is introduced to improve the ability of effectively breaking away from the local optimum. By adding the mutation operation to the basic PSO algorithm, the proposed algorithm can maintain the characteristic of fast speed. Finally, the significant performances in quality of the optimal solutions, the global search ability and convergence speed of algorithm proposed in this paper are validated by optimizing four benchmark functions.