首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于改进RBF神经网络的混沌时间序列预测
引用本文:郭兰平,俞建宁,张旭东,漆玉娟,张建刚.基于改进RBF神经网络的混沌时间序列预测[J].云南民族大学学报(自然科学版),2011,20(1):63-70.
作者姓名:郭兰平  俞建宁  张旭东  漆玉娟  张建刚
作者单位:兰州交通大学,数理与软件工程学院,甘肃,兰州,730070
基金项目:国家自然科学基金,甘肃省自然科学基金,兰州交通大学大学生科技创新基金
摘    要:基于RBF神经网络与相空间重构理论,对网络预测模型进行改进,并以Lorenz动力系统产生的混沌时间序列作为研究对象,建立预测模型并对其进行数值仿真.实验结果表明,基于改进RBF神经网络与相空间重构理论的混沌时间序列预测方法比BP、RBF神经网络模型的预测精度高、误差小、性能优越,改进方法可行、有效.

关 键 词:RBF神经网络  相空间重构  嵌入维数  延迟时间  混沌时间序列

Chaotic Time Series Forecasting Model Based on the Improved RBFNN
GUO Lan-ping,YU Jian-ning,ZHANG Xu-dong,QI Yu-juan,ZHANG Jian-gang.Chaotic Time Series Forecasting Model Based on the Improved RBFNN[J].Journal of Yunnan Nationalities University:Natural Sciences Edition,2011,20(1):63-70.
Authors:GUO Lan-ping  YU Jian-ning  ZHANG Xu-dong  QI Yu-juan  ZHANG Jian-gang
Institution:GUO Lan-ping,YU Jian-ning,ZHANG Xu-dong,QI Yu-juan,ZHANG Jian-gang(School of Mathematics,Physics & Software Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)
Abstract:Based on RBF neural network and theory of phase space reconstruction,this research improved the network prediction model,and took the chaotic time series that was generated by Lorenz dynamical system as the research object.Prediction models were built and their numerical simulation was carried out.Experimental results show that with the improved RBF neural network and the theory of phase-space reconstruction of chaotic time series forecasting model with higher predictive precision,this one has fewer errors ...
Keywords:RBF neural network  phase space reconstruction  embedding dimension  delay time  chaotic time series  
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号