滤波辨识(13):类多变量方程误差ARMA系统的滤波递阶广义增广迭代参数辨识
Filtering Identification.Part M:Filtering-Based Hierarchical Generalized Extended Iterative Parameter Identification for Multivariable Equation-Error ARMA-Like Systems
丁锋 1万立娟 2栾小丽 3徐玲 4刘喜梅2
作者信息
- 1. 江南大学 物联网工程学院,江苏无锡 214122;青岛科技大学 自动化与电子工程学院,山东青岛 266061
- 2. 青岛科技大学 自动化与电子工程学院,山东青岛 266061
- 3. 江南大学 物联网工程学院,江苏无锡 214122
- 4. 常州大学 微电子与控制工程学院,江苏常州 213159
- 折叠
摘要
针对类多变量方程误差自回归滑动平均(M-EEARMA-like)系统,利用滤波辨识理念和递阶辨识原理,研究和提出了滤波递阶广义增广梯度迭代辨识方法、滤波递阶多新息广义增广迭代梯度辨识方法、滤波递阶广义增广最小二乘迭代辨识方法、滤波递阶多新息广义增广最小二乘迭代辨识方法.这些滤波递阶广义增广迭代参数辨识方法可以推广到其他有色噪声干扰下的线性和非线性多变量方程误差随机系统中.
Abstract
For multivariable equation-error autoregressive moving average-like(M-EEAR-MA-like)models,this paper investigates and proposes filtered hierarchical generalized ex-tended gradient-based iterative identification methods,filtered hierarchical multi-innovation generalized extended gradient-based iterative identification methods,filtered hierarchical generalized extended least squares-based iterative identification methods,and filtered hierar-chical multi-innovation generalized extended least squares-based iterative identification meth-ods by means of the filtering identification idea and the hierarchical identification principle from available input-output data.These filtered hierarchical generalized extended iterative i-dentification methods can be extended to other linear and nonlinear multivariable stochastic systems with colored noises.
关键词
参数估计/迭代辨识/辅助模型辨识/多新息辨识/递阶辨识/滤波辨识/最小二乘/多变量系统Key words
parameter estimation/iterative identification/auxiliary model identification/multi-innovation identification/hierarchical identification/filtering identification/least squares/multivariable system引用本文复制引用
出版年
2024
NETL
NSTL
万方数据