基于神经网络的混沌时间序列建模及预测

该文从相空间重构理论出发,讨论了基于神经网络的混沌时间序列建模及预测方法,并以Logistic方程产生的混沌时间序列作为研究对象,采用BP和RBF两种神经网络分别对其进行了仿真分析,实验结果表明:最大Lyapunov指数越大,可预测步长越短;基于RBF网络的混沌时间序列建模及预测效果优于BP网络。

The Modeling and Forecasting of Chaotic Time Series Based on Neural Network

With the analysis of the technology of phase space reconstruction,a modeling and forecasting technique based on neural network for chaotic time series is presented in this paper.Simulation experiments of chaotic time series produced by Logistic equation are proceeded by a BP neural network or by a RBF neural network.The experimental results show that as the largest Lyapunov exponent increases,the horizon of predictability decreases and the performance of the modeling and forecasting method of chaotic time series by the RBF neural network is superior to the performance by the BP neural network.