Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (12): 134-137.DOI: 10.3778/j.issn.1002-8331.2009.12.044

• 数据库、信号与信息处理 • Previous Articles     Next Articles

Chaotic time series prediction based on ANFIS with adaptive mutation differential evolution algorithm

LI Mu1,2,HE Yi-gang2,ZHOU Shao-wu1,TAN Wen1   

  1. 1.School of Information and Electrical Engineering,Hunan University of Science and Technology,Xiangtan,Hunan 411201,China
    2.College of Electrical and Information Engineering,Hunan University,Changsha 410082,China
  • Received:2008-03-03 Revised:2008-05-26 Online:2009-04-21 Published:2009-04-21
  • Contact: LI Mu

混沌时间序列的自适应变异差分进化ANFIS预测

李 目1,2,何怡刚2,周少武1,谭 文1   

  1. 1.湖南科技大学 信息与电气工程学院,湖南 湘潭 411201
    2.湖南大学 电气与信息工程学院,长沙 410082
  • 通讯作者: 李 目

Abstract: A new method of chaotic time series prediction based on Adaptive Neuro Fuzzy Inference System(ANFIS) with Adaptive Mutation Differential Evolution(AMDE) algorithm was proposed.In the proposed method,the structure parameters of ANFIS were optimized by the hybrid algorithm which combined AMDE algorithm with least square method.The initial parameters of ANFIS model were tuned by differential evolution algorithm and then the final parameters of the model were optimized by least square algorithm.The hybrid algorithm greatly raises the convergence speed of network parameters identification and the global convergence of the system.It is demonstrated that the presented method is effective by the computer simulation.

摘要: 提出了一种基于自适应变异差分进化(AMDE)算法的ANFIS模型对混沌时间序列进行预测的方法,该方法采用自适应变异差分进化算法和最小二乘法相结合的混合学习算法对ANFIS网络结构参数进行优化设计,利用差分进化算法的全局寻优能力对ANFIS网络前件参数进行优化,而网络的结论参数采用最小二乘法优化,混合学习算法提高了网络参数辨识的收敛速度和系统的全局收敛性,仿真实验结果表明了该方法的有效性。