Computer Engineering and Applications ›› 2014, Vol. 50 ›› Issue (21): 8-13.

Previous Articles     Next Articles

Quantum-behaved particle swarm optimization method based on Q-learning

SHENG Xinyi1, SUN Jun2, ZHOU Di1, XU Wenbo2   

  1. 1.School of Digital Media, Jiangnan University, Wuxi, Jiangsu 214021, China
    2.School of Internet of Things, Jiangnan University, Wuxi, Jiangsu 214021, China
  • Online:2014-11-01 Published:2014-10-28

一种Q学习的量子粒子群优化方法

盛歆漪1,孙  俊2,周  頔1,须文波2   

  1. 1.江南大学 数字媒体学院,江苏 无锡 214021
    2.江南大学 物联网学院,江苏 无锡 214021

Abstract: Quantum-behaved Particle Swarm Optimization algorithm is analyzed; contraction-expansion coefficient and its control method are studied. To the different performance characteristics with different coefficients control strategies, a control method of coefficient with Q-learning is proposed. The proposed method can tune the coefficient adaptively and the whole optimization performance is increased. The comparison and analysis of results with the proposed method, constant coefficient control method, linear decreased coefficient control method and non-linear decreased coefficient control method based on CEC2005 benchmark function is given.

Key words: Particle Swarm Optimization(PSO) algorithm, Q-learning, parameter selection, quantum behavior

摘要: 分析了量子行为粒子群优化算法,着重研究了算法中的收缩扩张参数及其控制方法,针对不同的参数控制策略对算法性能的影响特点,提出将Q学习方法用于算法的参数控制策略,在算法搜索过程中能够自适应调整选择参数,提高算法的整体优化性能;并将改进后的Q学习量子粒子群算法与固定参数选择策略,线性下降参数控制策略和非线性下降参数控制策略方法通过CEC2005 benchmark测试函数进行了比较,对结果进行了分析。

关键词: 粒子群算法, Q学习, 参数选择, 量子行为