Computer Engineering and Applications ›› 2011, Vol. 47 ›› Issue (15): 46-48.

• 研究、探讨 • Previous Articles     Next Articles

Adaptive-grouping particle swarm algorithm

FU Qiang   

  1. College of Science and Technology,Ningbo University,Ningbo,Zhejiang 315212,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-05-21 Published:2011-05-21


符 强   

  1. 宁波大学 科学技术学院,浙江 宁波 315212

Abstract: Based on niche ideas and catastrophe theory,a new particle swarm optimization algorithm(AGPSO) is proposed in this paper,which can adaptively adjust population structure.The algorithm,after obtaining local optimal area,leaves only part of the particles to find local minima.While the other particles are dealt with by disaster,and are restrained in the remaining regions for new search.The experiment results of three classic benchmark functions show that the algorithm is a better method to solve the premature convergence problem,it can not only improve the convergence velocity and precision in the evolutionary,but also effectively enhance the global optimization power.

Key words: particle swarm algorithm, adaptive-grouping, niche, disaster, interval-constrained, population structure

摘要: 结合小生境思想及灾变原理,提出了一种动态调整种群结构的粒子群算法(AGPSO)。该算法在获取局部最优区域后只留下部分粒子寻找局部最优点,同时将其他粒子进行灾变处理,然后约束在剩余区域进行新最优区域搜索,这样既达到了快速局部收敛的目的,同时又增加了粒子种群的多样性,较好地解决了早熟收敛的问题。通过典型优化函数的仿真实验验证了该算法的有效性。

关键词: 粒子群算法, 自适应分群, 小生境, 灾变, 域约束, 种群结构