Subspace-Based Optimal Iv Method for Closed-Loop System Identification

Gilson, Marion; Mercère, Guillaume

doi:10.3182/20060329-3-au-2901.00171

Cited by 11 publications

(17 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…If this is not satisfied, the subspace algorithms will in general return biased system parameters. Recently, several closed-loop subspace methods try to account for this Verhagen (1993); Ljung and McKelvey (1996); Qin and Ljung (2003); Jansson (2003); Qin (2006); Katayama et al (2005); Wang and Qin (2006); Gilson and Mercere (2006);Jansson (2005), and consistency analysis of different closed-loop subspace methods is given in Lin et al (2004) and Chiuso and Picci (2005).…”

Section: Introductionmentioning

confidence: 99%

Effect of Varying Controller Parameters in Closed-Loop Subspace Identification

Bakke¹,

Johansen²,

Skogestad

2010

IFAC Proceedings Volumes

View full text Add to dashboard Cite

It is well known that subspace identification methods that assume open loop data without correlations between the input and noise, may give biased estimates when applied to closed loop data. The effect of the controller gain parameters on the quality of the identified model is studied when closed loop data are used. Several subspace identification methods (both open loop and closed loop methods), and different simulated data sets ranging from ideal 2 x 2 linear systems, to a fairly realistic nonlinear debutanizer process simulator, are considered. The results show that up to a point, higher controller gain during the identification experiment gives more accurate models than with lower controller gain, for both open and closed loop subspace identification methods. It is observed that the sensitivity to the controller gain is very small for the closed loop sub-space method tested for the ideal cases when its assumptions are satisfied. An explaination for this is that in this case there will be no bias, while the open loop methods may have a bias that depends on the controller parameters. Another interesting observation is that for the debutanizer example, the nonlinearities seem to lead to biased estimates also with the closed loop subspace method, and the choice of controller gain appears to be just as important as the use of a closed loop subspace identification method for the accuracy of the estimates.

show abstract

Section: Introductionmentioning

confidence: 99%

Effect of Varying Controller Parameters in Closed-Loop Subspace Identification

Bakke¹,

Johansen²,

Skogestad

2010

IFAC Proceedings Volumes

View full text Add to dashboard Cite

show abstract

“…This interest is due to the ability of the subspace approach in providing accurate state-space models for multivariable linear systems directly from input-output data [1,2,3,4,5]. Recently, system identification is focused on closed-loop system applications since there are many cases where open-loop experiments are impossible due to safety and stability consideration [2,6,7]. Closed-loop experiments are also necessary if the open-loop plant is unstable, or the feedback is an inherent mechanism of the systems.…”

Section: Introductionmentioning

confidence: 99%

“…This is due to the correlation between disturbances and the control signal, induced by loop, in which the ordinary subspace methods failed to solve. However, with special treatment, now the subspace methods are also able to identify the closed-loop system (See for some published examples in [6,7,8,9,10]). …”

Section: Introductionmentioning

confidence: 99%

Performance Measure of Some Subspace-Based Methods for Closed-Loop System Identification

Aziz

Mohd‐Mokhtar

2010

2010 Second International Conference on Computational Intelligence, Modelling and Simulation

View full text Add to dashboard Cite

-This paper presents some of the methods under subspace-based family to perform closed-loop system identification. Three methods have been observed; those are the ORT method, MOESP method and CCA method. The performance of these models is evaluated by identifying the same experimental systems. Three performance evaluation tests according to mean square errors, variance accounted for and best fit are used to verify the accuracy of the models to identify the given systems. The noise model is also introduced in order to see whether the developed model can improve the overall identification performance or not. Based on the simulation results, some justifications are made as to conclude the efficacy of the observed models to identify a closed-loop data system.

show abstract

“…This example is also adopted in [24], [20], [13], [12], [21] The exogenous input r(t) is a gaussian white noise sequence with variance 1. The innovation e(t) is a gaussian white noise with variance 1/9 and the noise model is given by…”

Section: Numerical Examplementioning

confidence: 99%

“…• In the direct approaches the identification is performed as in an usual open loop context up to a suitable data processing ( [25], [7], [30], [15], [23], [24], [16], [2], [3], [13], [14], [4], [39], [12]); • The indirect approaches are mainly based on an open loop identification of the control system sensitivity function using the system output and an external excitation input ( [36]- [37], [27], [29], [28]); • The joint input-ouput approaches use the system inputoutput behavior together with an external excitation input ( [38], [17], [19], [21], [22]). …”

Section: Introductionmentioning

confidence: 99%

An indirect closed loop subspace identification method

Pouliquen

Gehan²,

Pigeon³

2010

49th IEEE Conference on Decision and Control (CDC)

View full text Add to dashboard Cite

Abstract-In this paper, we consider a closed-loop subspace identification problem. An indirect method is developed using exogenous input and knowledge of a part of the controller impulse response. The idea is to extract dynamic of the plant from dynamic of the closed loop system. Two main result allows this double estimation. Only the deterministic behavior of the plant is considered in this paper. A simulation example is given to illustrate the performances of the present method.

show abstract

Subspace-Based Optimal Iv Method for Closed-Loop System Identification

Cited by 11 publications

References 18 publications

Effect of Varying Controller Parameters in Closed-Loop Subspace Identification

Effect of Varying Controller Parameters in Closed-Loop Subspace Identification

Performance Measure of Some Subspace-Based Methods for Closed-Loop System Identification

An indirect closed loop subspace identification method

Contact Info

Product

Resources

About