全文下载： 
202106001.pdf
文章编号： 1672-6987（2021）06-0001-20； DOI： 10.16351/j.1672-6987.2021.06.001
丁锋1,2, 刘喜梅2(1.江南大学 物联网工程学院, 江苏 无锡 214122; 2.青岛科技大学 自动化与电子工程学院, 山东 青岛 266061)
摘要： 为了处理损失数据系统和稀少量测数据系统辨识，引入了变间隔概念。以线性回归系统为例，利用递阶辨识原理，提出了一些变间隔递阶递推辨识方法，包括变间隔递阶随机梯度辨识方法、变间隔递阶多新息随机梯度辨识方法、变间隔递阶递推梯度辨识方法、变间隔递阶多新息递推梯度辨识方法、变间隔递阶最小二乘辨识方法、变间隔递阶多新息最小二乘辨识方法等。这些变间隔递阶辨识方法可以推广到有色噪声干扰下的线性和非线性随机系统中。
关键词： 参数估计; 递推辨识; 多新息辨识; 递阶辨识; 最小二乘; 线性回归模型
中图分类号： TP 273文献标志码： A
引用格式： 丁锋, 刘喜梅. 传递函数辨识(23): 线性回归系统的变间隔递阶递推参数估计［J］. 青岛科技大学学报(自然科学版), 2021, 42(6): 1-20.
DING Feng, LIU Ximei. Transfer function identification. Part W: Interaval-varying hierarchical recursive parameter estimation for linear regressive models［J］. Journal of Qingdao University of Science and Technology (Natural Science Edition), 2021, 42(6): 1-20.
Transfer Function Identification. Part W: Interval-Varying Hierarchical
Recursive Parameter Estimation for Linear Regressive Models
DING Feng1,2, LIU Ximei2
(1.School of Internet of Things Engineering, Jiangnan University, Wuxi 214122, China;
2.College of Automation and Electronic Engineering, Qingdao University of Science and Technology, Qingdao 266061, China)
Abstract: In order to deal with the identification problems of missing-data systems or scarce-data systems, the interval-varying concept is introduced.Taking the linear regressive systems as an example, this paper presents some interval-varying hierarchical recursive identification methods, including the interval-varying hierarchical stochastic gradient identification methods, the interval-varying hierarchical multi-innovation stochastic gradient identification methods, the interval-varying hierarchical recursive gradient identification methods, the interval-varying hierarchical multi-innovation recursive gradient identification methods, the interval-varying hierarchical least squares identification methods, and the interval-varying hierarchical multi-innovation least squares identification methods, etc., based on the hierarchical identification principle.These interval-varying hierarchical identification methods can be extended to other linear and nonlinear stochastic systems with colored noises.
Key words: parameter estimation; recursive identification; multi-innovation identification; hierarchical identification; least squares; linear regressive model
收稿日期： 2021-10-21
基金项目： 国家自然科学基金项目(61873111，61472195).
作者简介： 丁锋(1963—)，男，博士，“泰山学者”特聘教授，博士生导师.
