全文下载: 202206001.pdf
文章编号: 1672-6987(2022)06-0001-10; DOI: 10.16351/j.1672-6987.2022.06.001
丁锋1,2, 刘海波1, 刘喜梅2(1. 青岛科技大学 自动化与电子工程学院, 山东 青岛 266061;
2. 江南大学 物联网工程学院, 江苏 无锡 214122)摘要: 基于系统的输入输出数据, 利用滤波辨识理念, 研究和提出了方程误差自回归系统的滤波广义随机梯度辨识方法、滤波多新息广义随机梯度辨识方法、滤波多新息广义投影辨识方法、滤波递推广义梯度辨识方法、滤波多新息递推广义梯度辨识方法、滤波递推广义最小二乘辨识方法、滤波多新息广义最小二乘辨识方法。这些滤波广义辨识方法可以推广到其他有色噪声干扰下的线性多变量和非线性多变量随机系统中。
关键词: 参数估计; 递推辨识; 多新息辨识; 递阶辨识; 最小二乘; 随机系统
中图分类号: TP 273文献标志码: A
引用格式: 丁锋, 刘海波, 刘喜梅. 滤波辨识(3): 方程误差自回归系统的滤波递推广义参数估计[J]. 青岛科技大学学报(自然科学版), 2022, 43(6): 1-10.
DING Feng, LIU Haibo, LIU Ximei. Filtering identification. Part C: Filtering-based recursive generalized parameter estimation for equation-error autoregressive systems[J]. Journal of Qingdao University of Science and Technology(Natural Science Edition), 2022, 43(6): 1-10.
Filtering Identification. Part C: Filtering-Based Recursive Generalized
Parameter Estimation for Equation-Error Autoregressive Systems
DING Feng1,2, LIU Haibo1, LIU Ximei2
(1. College of Automation and Electronic Engineering, Qingdao University of Science and Technology, Qingdao 266061, China;
2. School of Internet of Things Engineering, Jiangnan University, Wuxi 214122, China)
Abstract: By using the input-output data and the filtering identification idea, this paper presents a filtering-based generalized stochastic gradient identification method, a filtering-based multi-innovation generalized stochastic gradient identification method, a filtering-based multi-innovation generalized projection identification method, a filtering-based recursive generalized gradient identification method, a filtering-based multi-innovation recursive generalized gradient identification method, a filtering-based recursive generalized least squares identification method, and a filtering-based multi-innovation recursive generalized least squares identification method for equation-error autoregressive systems. The proposed filtering-based recursive generalized identification methods can be generalized to other linear and nonlinear multivariable stochastic systems with colored noises.
Key words: parameter estimation; recursive identification; multi-innovation identification; hierarchical identification; least squares; stochastic system
收稿日期: 2022-10-20
基金项目: 国家自然科学基金项目(61472195).
作者简介: 丁锋(1963—),男,博士,“泰山学者”特聘教授,博士生导师.