General particle swarm optimization algorithm

Abstract

In order to make full use of the individual extreme information of each particle, and promote optimization performance of particle swarm optimization algorithm, a new particle swarm optimization algorithm is proposed and named as generalized particle swarm optimization (GPSO). GPSO algorithm modifies the standard PSO algorithm speed update formula, in the current particle, cognitive part of speed update formula introduces the rest particles individual extreme information, it makes every particle get n (particle swarm size) update speeds and n corresponding location update values, and GPSO chooses the particle which has optimal adaptive value as the final iteration particle. similar to the PSO algorithm, Global extremum of GPSO and individual extremum of particle are updated. the nature of the GPSO algorithm is analyzed, GPSO algorithm enhances the capacity of information exchange and sharing between the particles and has a downward compatibility for the PSO algorithm, which theoretically ensures that the GPSO algorithm has better search performance than the PSO algorithm. The simulation results show that the proposed algorithm is effective.