Algorithm design and experimental analysis for hybrid PSO based on VDGRASP

The intensification mechanism and diversification mechanism of Basic Particle Swarm Optimization (BPSO) algorithm were analyzed, and the precision problem of BPSO was discussed. Then a new hybrid PSO algorithm was set up, and the Variable Dimension Greedy Randomized Adaptive Search Procedure (VDGRASP) was designed to combine with BPSO. Two parameters of the hybrid PSO, SubDNum and PerturbScope, were analyzed through experiments, and the adaptive variations were designed for them. Three benchmarks were tested and show that the new algorithm is better than the BPSO with precision.