Particle Swarm Optimization Research Base on Quantum Q-Learning Behavior

Particle Swarm Optimization Research Base on Quantum Q-Learning Behavior

Lu Li (Information Center, Changsha Health Vocational College, Changsha, China) and Shuyue Wu (School of Information Science and Engineering, Hunan International Economics University, Changsha, China)
Copyright: © 2017 |Pages: 10
DOI: 10.4018/JITR.2017010103
OnDemand PDF Download:
$30.00
List Price: $37.50

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 is given based on CEC 2005 benchmark function.
Article Preview

1. Introduction

Particle Swarm Optimization(PSO) is a colony optimization algorithm which was proposed by Kennedy, who was inspired by the foraging behavior of birds (Kennedy & Eberhart, 1995). Because the method is simple in PSO algorithm, it has fast convergence, PSO algorithm has been widely used; but the algorithm itself is concerned, PSO is not a global optimization algorithm (Van den Bergh, 2001), many scholars have proposed many improved methods, some improvement effects have made (Zheng, Ma & Zhang, 2003; Clerc, 2004). Sun et al. had studied the intelligent population evolution in-depth based on the analysis of particle swarm optimization algorithm, quantum theory was introduced into a PSO algorithm, quantum-behaved particle swarm optimization algorithm was proposed with a global search capability (Quantum-behaved Particle Swarm Optimization, QPSO) (Sun, Feng & Xu, 2004; Sun, Xu & Feng, 2004). Since the proposed PSO algorithm, it has the simple calculation program, and it is easy to implement, and there are less control parameters, etc., it caused research and attention of many scholars in related fields at home and abroad (Goh, Tan, Liu et al., 2010; Omranpour, Ebadzadeh, Shirt et al., 2012), but also it has been applied to some practical problems (Chen, Sun & Ding, 2008; Chai, Sun, Cai et al., 2009; Omkar, Khandelwal, Ananth et al., 2009). QPSO algorithm has only parameter (contraction expansion factor), Sun et al., used a fixed parameter control strategy (Sun & Wu, 2012; Sun, Fang, Wu et al., 2012), later Fang proposed increases evolutionary numbers, the linear or nonlinear decreasing parameter control method were used, the simulation results showed that good improvement effect are achieved in most of the test functions (Fang, 2008). For particle contribution differences in QPSO algorithm, Xi et al. proposed weighting coefficient control algorithm and search process improvement methods, but also it achieved some results (Xi, Sun, & Xu, 2008).

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 10: 4 Issues (2017)
Volume 9: 4 Issues (2016)
Volume 8: 4 Issues (2015)
Volume 7: 4 Issues (2014)
Volume 6: 4 Issues (2013)
Volume 5: 4 Issues (2012)
Volume 4: 4 Issues (2011)
Volume 3: 4 Issues (2010)
Volume 2: 4 Issues (2009)
Volume 1: 4 Issues (2008)
View Complete Journal Contents Listing