Technical Briefs

Closed-Loop System Identification Based on Data Correlation

[+] Author and Article Information
Hassene Jammoussi

e-mail: hjammoussi@uh.edu

Karolos Grigoriadis

Department of Mechanical Engineering,
University of Houston,
Houston, TX 77204

Martin Books

Power Systems Controls,
Cummins Inc.,
Columbus, IN 47203

1Corresponding author.

Contributed by the Dynamic Systems Division of ASME for publication in the JOURNAL OF DYNAMIC SYSTEMS, MEASUREMENT, AND CONTROL. Manuscript received September 26, 2012; final manuscript received July 2, 2013; published online October 18, 2013. Assoc. Editor: Gregory Shaver.

J. Dyn. Sys., Meas., Control 136(1), 014507 (Oct 18, 2013) (10 pages) Paper No: DS-12-1319; doi: 10.1115/1.4025158 History: Received September 26, 2012; Revised July 02, 2013

A closed-loop system identification method is developed to estimate the parameters of a single input single output (SISO) linear time invariant system (LTI) operating within a feedback loop. The method uses the reference command in addition to the input–output data and establishes a correlation framework to structure the system. The correlation-based method is capable of delivering consistent estimates provided that the specific conditions on the signals are met. The method parallels the instrumental variables four step algorithm (IV4) and is comprised of three steps. First a model is estimated using cross correlation calculations between the reference input signal and the control and measured output signals. In the second step, a prefilter is identified to reduce estimation bias. In the final step, the prefilter, the instrumental variables and the measured signals are employed to estimate the final model. A consistency proof is provided for the proposed estimation process. The method is demonstrated on two examples. The first uses data collected from a diesel engine operation, and an open-loop model relating fueling to engine speed is sought. The identification process is complicated by the presence of nonmeasurable external torque disturbances and stochastic sensor noise. The second example uses data obtained from a time domain simulation of a closed-loop system where high levels of nonmeasured noise and disturbances were considered and a comparison with existing methods is made.

Copyright © 2014 by ASME
Your Session has timed out. Please sign back in to continue.


Brillinger, D., 1981, Time Series: Data Analysis and Theory, San Francisco, CA.
Sinha, N., and Kuszta, B., 1983, Modeling and Identification of Dynamic Systems, Van Nostrand Reinhold Company Inc., NewYork.
Katayama, T., 2005, Subspace Methods for System Identification, Springer, Berlin.
Van Den Hof, P., and Shrama, R., 1995, “Identification and Control-Closed-Loop Issues,” Automatica, 31, pp. 1751–1770. [CrossRef]
Zeng, J., and De Callafon, R., 2006, “Control Relevant Estimation of Plant and Disturbance Dynamics,” Automatica, 42, pp. 1951–1957. [CrossRef]
Gevers, M., and Ljung, L., 1986, “Optimal Experiment Designs with Respect to the Intended Model Application,” Automatica, 22(5), pp. 543–554. [CrossRef]
Gevers, M., 1993, Towards a Joint Design of Identification and Control?, “Essays on Control: Perspectives in the Theory and its Applications,” H. L.Trentelman and J. C.Willems, eds., Birkhauser, Boston, MA, pp. 111–151.
Gustavsson, I., Ljung, L., and Soderstrom, T., 1977, “Identification of Processes in Closed-Loop; Identifiability and Accuracy Aspects,” Automatica, 13, pp. 59–75. [CrossRef]
Forssell, U., and Ljung, L., 1999. “Closed-Loop Identification Revisited,” Automatica, 35, pp. 1215–1241. [CrossRef]
Ljung, L., 1999, System Identification: Theory for the User, 2nd ed., Prentice Hall, Saddle River, NJ.
Gevers, M., Ljung, L., and Van den Hof, P., 2001, “Asymptotic Variance Expressions for Closed-Loop Identification,” Automatica, 37, pp. 781–786. [CrossRef]
Soderstrom, T., and Stoica, P., 1989, System Identification, Prentice Hall, New-York.
Ljung, L., and Forssell, U., 1999, “An Alternative Motivation for the Indirect Approach to Closed-Loop Identification,” IEEE Trans. Autom. Control, 44(11), pp. 2206–2209. [CrossRef]
Akaike, H., 1968, “On the Use of a Linear Model for the Identification of Feedback Systems,” Ann. Inst. Stat. Math., 20(11), pp. 425–439. [CrossRef]
Soderstrom, T., and Stoica, P., 1981, “Comparison of Some Instrumental Variable Methods—Consistency and Accuracy Aspects,” Automatica, 17(1), pp. 101–115. [CrossRef]
Soderstrom, T., and Stoica, P., 1985, “Instrumental Variable Methods for Identification of Dynamic Systems,” Identification and System Parameter Estimation, York, UK, pp. 17–28.
Gilson, M., and Van Den Hof, P., 2005, “Instrumental Variable Methods for Closed-Loop System Identification,” Automatica, 41, pp. 241–249. [CrossRef]
Isermann, R., and Bauer, U., 1974, “Two-Step Process Identification With Correlation Analysis and Least-Squares Parameter Estimation,” ASME J. Dyn. Syst., Meas., Control, 96, pp. 426–432. [CrossRef]
Bendat, J., and Piersol, A., 2010, Random Data: Analysis and Measurement Procedures, Wiley Series in Probability and Statistics, Hoboken, NJ.
Ljung, L., and Soderstrom, T., 1987, Theory and Practice of Recursive Identification, The MIT Press; Cambridge, MA.
Gertler, J., 1998, Fault Detection and Diagnosis in Engineering Systems, Marcel Dekker, New York.
Guzzella, L., and Onder, C., 2004, Introduction to Modeling and Control of Internal Combustion Engine Systems, Springer-Verlag, Berlin.
Ljung, L., 2010, System Identification Toolbox 7: User's Guide, MathWorks, Incorporated, Natick, MA.


Grahic Jump Location
Fig. 1

Feedback control structure

Grahic Jump Location
Fig. 2

Least squares estimation for G1 (q−1)

Grahic Jump Location
Fig. 3

Estimation of filter F(q−1)

Grahic Jump Location
Fig. 4

Estimation of the final model G2 (q−1)

Grahic Jump Location
Fig. 5

Closed-loop engine system

Grahic Jump Location
Fig. 6

Data collected from an off-highway Cummins-QSB engine

Grahic Jump Location
Fig. 7

Predicted speed from proposed closed-loop identification versus measured speed

Grahic Jump Location
Fig. 8

Evolution of the coefficients a1 and b1 in terms of the iterations

Grahic Jump Location
Fig. 9

Collected data from the simulated closed-loop system

Grahic Jump Location
Fig. 10

Predicted output from proposed closed-loop identification vs. other conventional methods

Grahic Jump Location
Fig. 11

Bode plots of the exact model versus identified models from proposed IV3 algorithm and other conventional methods




Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In