适应性粒子群寻优算法

Adaptive particle swarm optimization algorithm

社会性的群体寻优是秩序与混沌之间的平衡,适应性微粒群寻优算法(APSO)是在标准PSO上添加反映适应性的随机项,并引入小概率因子,使微粒飞行到粒子群的中心,平衡秩序和随机两个行为.APSO算法的本质是在有序的决策中始终引入随机的、不可预测的决定,从而使得寻优的决策尽可能模拟社会性群体寻优的复杂行为.典型复杂函数优化的仿真结果表明,APSO算法具有较好的稳定性.

The social behavior of swarm is a balance between complete order and total chaos. Therefore, the level of randomness in the group is an important factor. Firstly, the concept of ＂chaos＂ is introduced to the particle swarm optimization（PSO） by adding a durative random item representing principle of adaptability of swarm intelligence. Then, particles with small probability will fly to the center of the swarm, which is introduced to balance the order and random behaviour. The essence of adaptive particle swarm optimization （APSO） is that the inscrutable decision on the rational behavior is introduced to order decision, and the complex behaviour of socical swarm is simulated. Experimental simulations show that the proposed method can improve the stability of convergence effectively.