| 引用本文: | 柴天佑 ， 寇新民 ， 罗公亮.动态过程辨识的一种BP鲁棒算法[J].控制理论与应用,1998,15(6):929~932.[点击复制] |
| KOU Xinmin ， LUO Gongliang，CHAI Tianyou.A Robust Back Propagation Algorithm for Identificationof Dynamical Systems[J].Control Theory & Applications,1998,15(6):929~932.[点击复制] |
|
| 动态过程辨识的一种BP鲁棒算法 |
| A Robust Back Propagation Algorithm for Identificationof Dynamical Systems |
| 摘要点击 1492 全文点击 651 投稿时间：1996-08-27 修订日期：1997-09-29 |
| 查看全文 查看/发表评论 下载PDF阅读器 HTML |
| DOI编号 |
| 1998,15(6):929-932 |
| 中文关键词 鲁棒算法 BP网 强干扰 系统辨识 |
| 英文关键词 robust algorithm,BP network gross errors,identification systems |
| 基金项目 |
|
| 中文摘要 |
| 针对BP算法在进行系统辨识时存在着速度慢、非平滑内插、受噪声影响很大、逼近精度不高，特别是对强干扰不具有鲁棒性等问题，提出一种BP网络的鲁棒算法，该算法直接利用样本点对样本点的分布特征进行估计，并采用带有损失因子的与误差分布有关的二次型能量函数，并用于动态系统辨识. 仿真结果表明了算法不仅对白噪声具有鲁棒性，而且对强干扰也具有鲁棒性. |
| 英文摘要 |
| A robust back propagation algorithm is proposed，which is aimed at solving problems that BP algorithm has some confronted when applied to the dynamic identification，such as slow convergence，non-smoothing interpolation，noise sensitivity and poor accuracy of identification，especially lack of robustness for strong interference. The algorithm uses samples directly to estimate distribution characteristics of sample points in which a quadratic energy function with lose factors relevant to error of distributions is used. The al-gorithm can be used in identification of dynamical systems，the simulation results showed the robustness for white noise as well as for gross errors. |