计算机工程与应用 ›› 2016, Vol. 52 ›› Issue (1): 141-145.

• 模式识别与人工智能 • 上一篇    下一篇

吸引子权重改变内嵌区域震荡搜索粒子群算法

朱  沛,范年柏   

  1. 湖南大学 信息科学与工程学院,长沙 410082
  • 出版日期:2016-01-01 发布日期:2015-12-30

Attractor weight change regional shock search embedded particle swarm optimization algorithm

ZHU Pei, FAN Nianbai   

  1. College of Information Science and Engineering, Hunan University, Changsha 410082, China
  • Online:2016-01-01 Published:2015-12-30

摘要: 针对粒子群算法搜索精度不高、搜索最优解较慢的问题,提出了一种改进的粒子群算法。该算法通过调整全局最优解和个体最优解,形成一个新的全局吸引子解指导粒子收敛,优化种群粒子来搜索解空间的最优值。再将优化方案融入到内嵌区域震荡搜索的粒子群算法(RSPSO)中,仿真结果表明,改进的粒子群算法在寻优能力及搜索精度方面都得到了进一步的提高。

关键词: 群体智能, 粒子群优化, 权重, 吸引子, 内嵌区域震荡搜索

Abstract: In order to solve the problem, which particle swarm optimization has low accuracy and slow speed in searching optimal, many scholars have proposed various methods which improve the optimization to make particle swarm algorithm more efficient. This paper presents an improved method by adjusting the global and individual optimal solution to form a new global attractor solution, optimizing the search for the optimal value of the particle populations in the solution space. Then embedds it into regional shock search embedded PSO(RSPSO), the simulation results show that searching ability and the search accuracy of the improved particle swarm optimization has been further improved.

Key words: swarm intelligence, particle swarm optimization, weight, attractors, regional shock search embedded