Fast Convergence Particle Swarm Optimization for Functions Optimization

Time to time, many researchers have suggested modifications to the standard particle swarm optimization to find good solutions faster than the evolutionary algorithms, but they could be possibly stuck in poor region or diverge to unstable situations. For overcoming such problems, this paper proposes new Fast Convergence Particle Swarm Optimization (FCPSO) approach based on balancing the diversity of location of individual particle by introducing a new parameter, particle mean dimension (Pmd) of all particles to improve the performance of PSO. The FCPSO method is tested with five benchmark functions by variable dimensions and fixed size population and compared with PSO & Constriction factor approach of PSO (CPSO). Finally, search performances of these methods on the benchmark functions are tested.