Grid-based adaptive particle swarm optimization
一种基于网格划分的自适应粒子群优化算法*

To improve the performance of PSO, this paper proposed aGAPSO, and developed a measurement method for warm diversity and a maximal diversity algorithm for the initial swarm (MDAIS). Divided the GAPSO into two phases. In the first phase, tuned the evolving direction of a particle adaptively according to the contribution to the diversity. In the second phase, tuned the inertia weight of a particle adaptively according to the contribution to the diversity. Used several classic benchmark functions to evaluate the GAPSO. The experimental results show that for continuous optimization problems, the GAPSO outperforms the classic PSO. The iteration times for finding the best solutions in the GAPSO decrease about 60% averagely while compared with that in the classic PSO.