Computer Engineering and Applications ›› 2011, Vol. 47 ›› Issue (10): 17-19.

• 博士论坛 • Previous Articles     Next Articles

Parameter optimization of fuzzy neural networks based on cultural quantum- behaved particle swarm optimization

ZHAO Jing1,2,SUN Jun1,XU Wenbo1   

  1. 1.School of Internet of Things Engineering,Jiangnan University,Wuxi,Jiangsu 214122,China
    2.Center of Modern Education Technology,Shandong Polytechnic University,Jinan 250353,China

  • Received:1900-01-01 Revised:1900-01-01 Online:2011-04-01 Published:2011-04-01


赵 晶1,2,孙 俊1,须文波1   

  1. 1.江南大学 物联网工程学院,江苏 无锡 214122
    2.山东轻工业学院 现代教育技术中心,济南 250353

Abstract: The parameter optimization of Fuzzy Neural Network(FNN) is function optimization.According to the low convergence precision of the existed algorithms,a hybrid Cultural Quantum-behaved Particle Swarm Optimization(C-QPSO) is introduced to train the parameter optimization of FNN.The results of experiment show the proposed technique is effective.

Key words: Fuzzy Neural Network(FNN), parameter optimization, Quantum-behaved Particle Swarm Optimization(QPSO), Cultural Algorithm(CA), chaotic time series

摘要: 模糊神经网络参数学习是一个函数优化问题。针对已有优化方法收敛精度不高的缺点,提出基于文化量子粒子群算法的模糊神经网络参数优化,并将其应用于混沌时间序列预测。仿真实例结果证实了该算法的优越性。

关键词: 模糊神经网络, 参数优化, 量子粒子群算法, 文化算法, 混沌时间序列