System Identification and Order Determination Using Canonical-forms

Matsubara, Mitsuru; Usui, Yusuke; Sugimoto, Sueo

doi:10.5687/sss.2005.102

Cited by 5 publications

(2 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…[11,10,15], T d (z) and its realization are obtained precisely. In proposed method, LQ decomposition that plays a key role in subspace method [10,5] is employed to do orthogonal decomposition of the joint I/O process, and performs the estimation of w d based on FIR model under the finite observable data {r 1 (t), r 2 (t), y(t), u(t), t = 1, 2, · · · , T }.…”

Section: Prefiltering Of the Joint I/o Process Based On Lq Decompositionmentioning

confidence: 99%

See 1 more Smart Citation

Prefiltering Based on Wold's Decomposition for Closed-loop System Identification

Matsubara¹

2011

Stochastic Systems Theory and its Applications (SSS)

Self Cite

View full text Add to dashboard Cite

In this paper, we propose a prefiltering method for closed-loop MIMO system identification in the framework of joint input-output approach. The purpose of proposed prefiltering method is to eliminate the noise process from the joint input-output process more precisely, and can be achieved by using the orthogonal decomposition of the joint input-output process based on Wold's decomposition. In our method, LQ decomposition and causal FIR model are used. Especially, the latter plays a key role to eliminate the noise process more precisely. In numerical experiments, the efficacy of the proposed method is verified by comparing some identification methods.

show abstract

Section: Prefiltering Of the Joint I/o Process Based On Lq Decompositionmentioning

confidence: 99%

“…To obtain T d (z) and its realization, PO-MOESP [10], ORT method [5] and maximum likelihood(ML) method [15] are used, where the model order is fixed as 8 in each method. Fig.…”

Section: Numerical Experimentsmentioning

confidence: 99%

Prefiltering Based on Wold's Decomposition for Closed-loop System Identification

Matsubara¹

2011

Stochastic Systems Theory and its Applications (SSS)

Self Cite

View full text Add to dashboard Cite

show abstract

Subspace system identification for CO<inf>2</inf> recovery processes

Dunia

Rochelle

Qin

2011

2011 IEEE International Symposium on Computer-Aided Control System Design (CACSD)

View full text Add to dashboard Cite

The development of amine scrubbing for coal and natural gas-fired power plants represents a key technology to reduce CO2 emissions. Among the strategies required to maximize CO2 capture during plant operations is the design of tailor-made dynamic models for optimal control. This paper presents a novel application of subspace system identification to a CO2 recovery plant, where major decision variables are considered to develop a simple state space model that can estimate more than sixty process outputs. This model demonstrates to have a great predictive potential, which opens opportunities for the implementation of robust predictive controllers that can quickly adjust to power plant load changes. I. INTRODUCTIONThe CO 2 capture process with amine solvent based absorption and stripping is the most significant industrial method for the removal of carbon dioxide from coal-and natural gas-fired power plants [1]. Several dynamic models have been developed for the CO 2 capture process [2] [3]. However, most of these models require reaction kinetic and thermodynamic properties of the species considered in the process. Therefore, to obtain the simulation results, rigorous property calculations are programmed in specialized software that runs in off-line hardware. Such a modeling implementation approach is useful for model based control design and optimization strategies. Nevertheless, the application of such models during process operations is limited because simulation convergence, time and computational resources are limited in industrial setups.A practical approach to obtain a dynamic model from a plant in operation is to use data-driven empirical models, where the model is made to match the process measurements. Among empirical models, the subspace system identification algorithms have been well accepted in industry not only because of their simplicity and robustness, but also because they provide state space form models that are very convenient for prediction, process monitoring and model based control [4] [5] [6]. This work applies subspace system identification to a CO 2 recovery pilot plant. The empirical model is developed by collecting data from more than seventy sensors and making typical operational changes to perturb the process. Such changes are expected to excite the different modes of operation to determine the number of states required to describe the dynamic response of the process.Although the models developed in this paper are calculated off-line, the equations can be easily incorporated into the

show abstract