| 引用本文: | 陆系群 ，余英林.一种自适应预测非平稳信号的新方法[J].控制理论与应用,1998,15(2):308~311.[点击复制] |
| LU Xiquan，YU Yinglin.A New Method of Adaptive Prediction of Nonstationary Signals[J].Control Theory and Technology,1998,15(2):308~311.[点击复制] |
|
| 一种自适应预测非平稳信号的新方法 |
| A New Method of Adaptive Prediction of Nonstationary Signals |
| 摘要点击 877 全文点击 359 投稿时间：1996-04-12 修订日期：1997-02-28 |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| DOI编号 |
| 1998,15(2):308-311 |
| 中文关键词 动态神经网络 非线性 非平稳性 时间序列 预测 |
| 英文关键词 dynamic neural network nonlinear nonstationary time series prediction |
| 基金项目 |
|
| 中文摘要 |
| 本文提出一种动态神经网络为非平稳信号作自适应单步预测，它由级联回归神经网络和抽头延时线组成。用它对非线性动态方程产生的时间序列作适应预测，实验结果表明，其效果远远超过了传统的前馈神经网络。 |
| 英文摘要 |
| In this paper, a dynamic neural network is proposed, which is used for the nonlinear adaptive prediction of nonstationary signals. The dynamic neural network consists of cascade –connected recurrent neural networks, and the linear section consists of a conventional tapped-delay-line filter. This kind of neural network is used for nonlinear adaptive prediction of nonlinear time series, generated by nonlinear dynamic equation. The experimental results show that it outperforms the traditional feedforward neural network. |