| 引用本文: | 王晓,韩崇昭,万百五.非线性系统辨识的改进双对角化最小二乘算法[J].控制理论与应用,1997,14(5):629~637.[点击复制] |
| WANG Xiao，HAN Chongzhao and WAN Baiwu.The Modified Bidiagonalization Least Squares Algorithm for Nonlinear System Identification[J].Control Theory and Technology,1997,14(5):629~637.[点击复制] |
|
| 非线性系统辨识的改进双对角化最小二乘算法 |
| The Modified Bidiagonalization Least Squares Algorithm for Nonlinear System Identification |
| 摘要点击 1201 全文点击 802 投稿时间：1995-12-18 修订日期：1996-08-19 |
| 查看全文 查看/发表评论 下载PDF阅读器 |
| DOI编号 |
| 1997,14(5):629-637 |
| 中文关键词 非线性系统 系统辨识 双对角化最小二乘 |
| 英文关键词 nonlinear system system identification bidiagonalization least squares |
| 基金项目 |
|
| 中文摘要 |
| 本文提出了一种新的有效的非线性系统最小二乘辨识算法—改进的双对角化最小二乘（ＭＢＬＳ）算法，在存在舍入误差的条件下，给出了算法的收敛性证明，事实上，算法的收敛性几何不受舍入误差的影响，算法是大范围数值稳定的.仿真结果说明了新算法的有效性. |
| 英文摘要 |
| A new effective least squares algorithm for nonlinear system identification has been proposed in this paper，which is referred to as the Modified Bidiagonalization Least Squares (MBLS)one. Under the condition that round off errors exist，a convergence proof on it has been given. In fact，it is numerically stable in a large category，and round-off errors have almost no influence to its convergence. The simulation results indicate its superiority. |