Bias compensation‐based parameter estimation for output error moving average systems
Bias compensation‐based parameter estimation for output error moving average systems
复制标题
DOI：
10.1002/acs.1266
复制
发表时间：
2011-12
影响因子：
3.1
通讯作者：
Jie Ding;F. Ding
中科院分区：
文献类型：
--
作者：
Jie Ding;F. Ding
在本文中考虑了带有移动平均值的输出误差模型。系统参数。参数估计是通过交互式估计噪声方差和噪声参数的。
Identification problems of output error models with moving average noises are considered in this paper. The least‐squares‐based parameter estimation is biased under the colored noises in outputs. Firstly, a bias compensation term is formulated to achieve the bias‐eliminated estimates of the system parameters. Secondly, the bias compensation term is determined by the unknown variance of the noise and the unknown noise model, thus based on the hierarchical identification principle, an unbiased parameter estimation is obtained by interactively estimating noise variance and noise parameters. Finally, the estimated bias compensation term is added to the biased parameter estimates. The simulation examples confirm the effectiveness of the proposed algorithm. Copyright © 2011 John Wiley & Sons, Ltd.