小波网络在带噪声的混沌时间序列预测中的研究

In the process of modeling chaotic time series, noise will weaken the generalization ability of neural network model, a forecast framework is proposed to solve this problem. Wavelet denoise is introduced in the preprocessing stage to depress the influence of noise, then embedding dimension and delay time are determined by phase space reconstruction theory to choose the best structure of hybrid wavelet neural network, finally the hybrid wavelet neural network model is trained by both genetic algorithm and back-propagation algorithm. Experiment on Mackey-Glass chaotic time series validates the framework.