基于广义泊松矩泛函的闭环子空间辨识

doi:10.12305/j.issn.1001-506X.2024.06.27

系统工程与电子技术 ›› 2024, Vol. 46 ›› Issue (6): 2092-2098.doi: 10.12305/j.issn.1001-506X.2024.06.27

• 制导、导航与控制 • 上一篇

基于广义泊松矩泛函的闭环子空间辨识

于淼, 王友谊, 魏永涛

东北大学秦皇岛分校控制工程学院, 河北秦皇岛 066004

收稿日期:2023-07-20 出版日期:2024-05-25 发布日期:2024-06-04
通讯作者: 于淼
作者简介:于淼(1986—), 女, 讲师, 博士, 主要研究方向为系统辨识、预测控制
王友谊(1998—), 男, 硕士研究生, 主要研究方向为系统辨识
魏永涛(1980—), 男, 副教授, 博士, 主要研究方向为智能系统辨识与优化
基金资助:
国家自然科学基金(62003082);河北省自然科学基金(F2021501018);河北省教育厅科学技术研究项目(ZD2022148)

Closed-loop subspace identification via generalized Possion moment functionals

Miao YU, Youyi WANG, Yongtao WEI

School of Control Engineering, Northeastern University at Qinhuangdao, Qinhuangdao 066004, China

Received:2023-07-20 Online:2024-05-25 Published:2024-06-04
Contact: Miao YU

摘要/Abstract

摘要：

针对连续系统在闭环条件下进行辨识, 由于将来输入和噪声的相关性, 会导致辨识方法产生有偏估计结果的问题, 提出一种基于广义泊松矩泛函的闭环子空间辨识算法。首先, 采用广义泊松矩泛函变换得到输入输出信号的滤波器模型, 从而得到连续系统的输入输出矩阵方程。然后, 利用故障诊断领域中的等价空间代替子空间辨识过程中的观测空间进行辨识。最后, 为了解决闭环辨识中由反馈控制器产生的估计偏差辨识问题, 采用主成分分析法和辅助变量法, 达到辨识系统一致性估计的目的。仿真结果验证了所提方法的有效性和精确性。

关键词: 子空间辨识, 闭环辨识, 广义泊松矩泛函, 等价空间

Abstract:

To solve the problem of biased estimation due to the correlation between future input and noise in the identification process of continuous system, a closed-loop subspace identification algorithm utilizing generalized Possion moment functionals is presented. Firstly, the filter model of the input-output signals is obtained by generalized Possion moment functionals transformation, and then the input-output matrix equation of the continuous systems is obtained. Secondly, instead of using the observable subspaces in the process of subspace identification, this paper focus on the parity space which is commonly employed in fault detection. Finally, the system is estimated consistently by the principal component analysis and instrumental variable method, which solves the identification problems of biased results for the system operates in closed-loop identification due to the feedback controller. The effectivity and accuracy of the proposed method are verified by the simulation results.

Key words: subspace identification, closed-loop identification, generalized Possion moment functionals (GPMF), parity space

中图分类号:

N945.14

于淼, 王友谊, 魏永涛. 基于广义泊松矩泛函的闭环子空间辨识[J]. 系统工程与电子技术, 2024, 46(6): 2092-2098.

Miao YU, Youyi WANG, Yongtao WEI. Closed-loop subspace identification via generalized Possion moment functionals[J]. Systems Engineering and Electronics, 2024, 46(6): 2092-2098.

图/表 4

参考文献 31

1	HUANG C , MAJUMDAR A . A combined invariant-subspace and subspace identification method for continuous-time state-space models using slowly sampled multi-sine-wave data[J]. Automatica, 2022, 140, 110261. doi: 10.1016/j.automatica.2022.110261
2	庞志雅, 马志赛, 丁千. 基于神经网络和子空间的非线性系统载荷识别[J]. 振动、测试与诊断, 2022, 42 (5): 877- 885.
	PANG Z Y , MA Z S , DING Q . Force identification for nonlinear systems based on neural network and subspace method[J]. Journal of Vibration, Measurement & Diagnosis, 2022, 42 (5): 877- 885.
3	VARANASI S K , JAMPANA P . Nuclear norm subspace identification of continuous time state-space models with missing outputs[J]. Control Engineering Practice, 2020, 95, 104239. doi: 10.1016/j.conengprac.2019.104239
4	LI K , LUO H , YIN S , et al. A novel bias-eliminated subspace identification approach for closed-loop systems[J]. IEEE Trans.on Industrial Electronics, 2021, 68 (6): 5167- 5205.
5	YU M , GUO G , LIU J C , et al. Closed-loop time-varying continuous-time recursive subspace-based prediction via principle angles rotation[J]. ISA Transactions, 2022, 122, 135- 145. doi: 10.1016/j.isatra.2021.04.047
6	GARNIER H , WANG L . Identification of continuous time models from sampled data[M]. London: Springer, 2008.
7	GARNIER H , MENSLER M , RICHARD A . Continuous-time model identification from sampled data: implementation issues and performance evaluation[J]. International Journal of Control, 2003, 76 (13): 1337- 1357. doi: 10.1080/0020717031000149636
8	BASTOGNE T , GARNIER H , SIBILLE P . A PMF-based subspace method for continuous time model identification: application to a multivariable winding process[J]. International Journal of Control, 2001, 74 (2): 118- 132. doi: 10.1080/00207170150203471
9	VARANASI S K , JAMPANA P . Nuclear norm subspace identification of continuous time state-space models with missing outputs[J]. Control Engineering Practice, 2020, 95 (7): 104239.
10	DAI M X , YANG X M . Continuous-time system identification with nuclear norm minimization and GPMF-based subspace method[J]. IEEE/CAA Journal of Automatica Sinica, 2016, 3 (2): 184- 191. doi: 10.1109/JAS.2016.7451106
11	PRASADA R G , SAHA D C , RAO T M , et al. A microprocessor-based system for on-line parameter identification in continuous dynamical systems[J]. IEEE Trans.on Industrial Electronics, 1982, 29 (3): 197- 201.
12	GARNIER H, SIBILLE P, MENSLER M, et al. Pilot crane identification and control in presence of friction[C]//Proc. of the 13th IFAC World Congresss, 1996: 107-112.
13	HOUTZAGER I , WINGERDEN J W V , VERHAEGEN M . Recursive predictor-based subspace identification with application to the real-time closed-loop tracking of flutters[J]. IEEE Trans.on Automatic Control, 2012, 20 (4): 934- 949.
14	HUANG B , DING S X , QIN S J . Closed-loop subspace identification: an orthogonal projection approach[J]. Journal of Process Control, 2005, 15 (1): 53- 66. doi: 10.1016/j.jprocont.2004.04.007
15	WANG J , QIN S J . Closed-loop subspace identification using the parity space[J]. Automatica, 2006, 42, 315- 320. doi: 10.1016/j.automatica.2005.09.012
16	KATAYAMA T , TANAKA H . An approach to closed-loop subspace identification by orthogonal decomposition[J]. Automatica, 2007, 43, 1623- 1630. doi: 10.1016/j.automatica.2007.02.011
17	WINGERDEN J W V , VERHAEGEN M . Subspace identification of bilinear and LPV systems for open-and closed-loop data[J]. Automatica, 2009, 45, 372- 381. doi: 10.1016/j.automatica.2008.08.015
18	CHIUSO A . On the asymptotic properties of closed-Loop CCA-type subspace algorithms: equivalence results and role of the future horizon[J]. IEEE Trans.on Automatic Control, 2010, 55 (3): 634- 649. doi: 10.1109/TAC.2009.2039239
19	BERGAMASCO M , LOVERA M . Continuous-time predictor-based subspace identification using Laguerre filters[J]. IET Control Theory & Applications, 2011, 5 (7): 856- 867.
20	ZHANG L , ZHOU D H , ZHONG M Y , et al. Improved closed-loop subspace identification based on principal component analysis and prior information[J]. Journal of Process Control, 2019, 80, 235- 246. doi: 10.1016/j.jprocont.2019.06.001
21	GUNES B , WINGERDEN J W V , VERHAEGEN M . Tensor networks for MIMO LPV system identification[J]. International Journal of Control, 2020, 93 (4): 797- 811. doi: 10.1080/00207179.2018.1501515
22	HU Y S , JIANG Y F , CALLAFON R A D . Variance reduction in covariance based realization algorithm with application to closed-loop data[J]. Automatica, 2020, 113, 108683. doi: 10.1016/j.automatica.2019.108683
23	夏悠然, 管军, 易文俊. 基于改进粒子群优化极限学习机的弹丸参数辨识[J]. 系统工程与电子技术, 2023, 45 (2): 521- 529.
	XIA Y R , GUAN J , YI W J . Projectile parameter identification: extreme learning machine optimized by improved particle swarm[J]. Systems Engineering and Electronics, 2023, 45 (2): 521- 529.
24	YU M , LIU J C , GUO G , et al. Recursive subspace identification of continuous-time systems using generalized Poisson moment functionals[J]. Circuits, Systems, and Signal Processing, 2022, 41, 1848- 1868. doi: 10.1007/s00034-021-01871-x
25	王永刚, 毛博年, 高东. 一种补偿t-OPT噪声的冗余捷联惯组故障检测方法[J]. 航空控制, 2023, 41 (3): 26- 32.
	WANG Y G , MAO B N , GAO D . SINS fault detection based on compensation t-OPT noise[J]. Aerospace Control, 2023, 41 (3): 26- 32.
26	JAHANDARI S , MATERASSI D . Sufficient and necessary graphical conditions for MISO identification in networks with observational data[J]. IEEE Trans.on Automatic Control, 2022, 67 (11): 5932- 5947. doi: 10.1109/TAC.2021.3130885
27	SZENTPETERI S , CSAJI B C . Non-asymptotic state-space identification of closed-loop stochastic linear systems using instrumental variables[J]. Systems & Control Letters, 2023, 178, 105565.
28	SUN S Y , YANG B , ZHANG Q L , et al. A microprocessor-based system for on-line parameter identification in continuous dynamical systems[J]. Mechanical Systems and Signal Processing, 2023, 197, 110326. doi: 10.1016/j.ymssp.2023.110326
29	张伟, 许爱强, 平殿发. 基于稀疏化核方法的非线性动态系统在线辨识[J]. 系统工程与电子技术, 2017, 39 (1): 223- 230.
	ZHANG W , XU A Q , PING D F . Nonlinear system online identification based on kernel sparse learning algorithm with adaptive regulation factor[J]. Systems Engineering and Electronics, 2017, 39 (1): 223- 230.
30	WANG J , QIN S J . A new subspace identification approach based on principal component analysis[J]. Journal of Process Control, 2002, 12, 841- 855. doi: 10.1016/S0959-1524(02)00016-1
31	GU S W , CHEN J H , XIE L . Automatic segmentation of batch processes into multi-local state-space models for fault detection[J]. Chemical Engineering Science, 2023, 267, 118274. doi: 10.1016/j.ces.2022.118274

CSID-GPMF		PBSIDw
极点偏差	极点方差	极点偏差	极点方差
0.002 4	1.014 5×10^-5	0.003 7	1.252 9×10^-5
0.003 2	1.102 9×10^-5	0.005 2	1.761 7×10^-5
0.002 6	1.035 8×10^-5	0.012 4	2.014 1×10^-5

[1]	窦立谦, 宗群, 刘文静. 面向控制的系统辨识研究进展[J]. Journal of Systems Engineering and Electronics, 2009, 31(01): 158-164.
[2]	窦立谦, 宗群, 刘文静. 面向控制的系统辨识研究进展[J]. Journal of Systems Engineering and Electronics, 2009, 31(1): 158-164.

基于广义泊松矩泛函的闭环子空间辨识

Closed-loop subspace identification via generalized Possion moment functionals

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

图/表 4

参考文献 31

相关文章 2

编辑推荐

Metrics

本文评价