摘要
针对非线性时间序列，建立了小波神经网络预测模型，通过计算小波分解和小波级数，达到最优的逼近效果．经实例验证，该方法能有效地提高预测精度。
In this paper, we build a wavelet neural network prediction model of time series. By computing the wavelet resolution and wavelet series, the best approximation effect is reached. It can be seen from the example that this method can effectively improve the prediction accuracy. The intrinsic defects of artificial neural network model are avoided.
出处
《系统工程理论与实践》
EI
CSCD
北大核心
1999年第5期89-92,共4页
Systems Engineering-Theory & Practice
基金
中华电力教育基金
关键词
小波
神经网络
时间序列
预测模型
系统工程
wavelet neural network
wavelet transform
nonlinear time series
prediction model
