首页 | 本学科首页   官方微博 | 高级检索  
     

增广递推最小二乘辨识精度的研究
引用本文:杨艺,张云生. 增广递推最小二乘辨识精度的研究[J]. 昆明理工大学学报(自然科学版), 2009, 34(6): 49-52. DOI: 10.3969/j.issn.1007-855x.2009.06.010
作者姓名:杨艺  张云生
作者单位:昆明理工大学,信息工程与自动化学院,云南,昆明,650051;昆明理工大学,信息工程与自动化学院,云南,昆明,650051
摘    要:以带有有色噪声的CARMA模型为例,分析了在干扰项参数辨识当中,误差较大的原因以及影响辨识误差的因素,并利用二次辨识的思想提出了一种提高辨识精度的方法。通过此方法并结合一个具体实例,给出了Matlab仿真辨识的结果。仿真结果表明,经过二次辨识的参数更接近真实值。

关 键 词:参数估计  最小二乘  二次辨识

The Study of Extended Least-Squares Identification Precision
YANG Yi,ZHANG Yun-sheng. The Study of Extended Least-Squares Identification Precision[J]. Journal of Kunming University of Science and Technology(Natural Science Edition), 2009, 34(6): 49-52. DOI: 10.3969/j.issn.1007-855x.2009.06.010
Authors:YANG Yi  ZHANG Yun-sheng
Affiliation:( Faculty of Information Engineering and Automation, Kunrning University of Science and Technology, Kunming 650051, China)
Abstract:The paper uses CARMA model as an example, and analyses the reason of error of noise disturbance parameter identification and the factor that influences identification error. Then introduce a method to improve the precision of identification based on the idea of twice identification. Besides, the paper makes use of the method in an example and gives Matlab emulation identification results. The results display that the parameters are more close to real value after twice identification.
Keywords:parameter identification  least - squares  twice identification
本文献已被 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号