| 时变系统辨识的多新息方法 |
| |
| 引用本文： | 丁锋,谢新民,方崇智.时变系统辨识的多新息方法[J].自动化学报,1996,22(1):85-91. |
| |
| 作者姓名： | 丁锋 谢新民 方崇智 |
| |
| 作者单位： | 1.清华大学自动化系 |
| |
| 摘 要： | 推广了估计时不变参数的单新息修正技术,提出了多新息辨识方法.该方法可以抑制坏
数据对参数估计的影响,具有较强的鲁棒性.分析表明多新息方法可以跟踪时变参数,计算
量也较遗忘因子最小二乘法和卡尔曼(Kalman)滤波算法要小.仿真结果说明多新息算法估
计系统参数是有效的.
|
| 关 键 词： | 参数估计 时变系统 收敛性 多新息技术 |
| 收稿时间： | 1993-7-26 |
Multi-Innovation Identification Method for Time-Varying Systems |
| |
| Ding Feng,Xie Xinmin,Fang Chongzhi.Multi-Innovation Identification Method for Time-Varying Systems[J].Acta Automatica Sinica,1996,22(1):85-91. |
| |
| Authors: | Ding Feng Xie Xinmin Fang Chongzhi |
| |
| Affiliation: | 1.Department of Automation,Tsinghua University,Beijing |
| |
| Abstract: | In this paper, the single innovation modification technique of estimating time-invariant parameters is extended, and the multi-innovation identification method is presented. This method may overcome the effect of bad data on the parameter estimation. It has stronger robustness and may track time-varying parameters. its computational burden is less than forgetting factor least squares algorithm and Kalman filter algorithm. The simulation results indicate that the multi-innovation algorithm works quite well. |
| |
| Keywords: | Parameter estimation time-varying system convergence multi-innovation technique |
| 本文献已被 CNKI 维普 等数据库收录！ |
| 点击此处可从《自动化学报》浏览原始摘要信息 |
|
点击此处可从《自动化学报》下载全文 |
|
