UDC 519.711
1 Space Research Institute of the National Academy of Sciences of Ukraine and State Space Agency of Ukraine, Kyiv, Ukraine
 v.f.gubarev@gmail.com
|
2 Institute of Applied System Analysis NTUU “Igor Sikorsky Kyiv Polytechnic Institute,” Kyiv, Ukraine
ipsa@kpi.ua
|
3 Institute of Applied System Analysis NTUU “Igor Sikorsky Kyiv Polytechnic Institute,” Kyiv, Ukraine
yuriy.milyavsky@gmail.com
|
|
METHODS OF FINDING REGULARIZED SOLUTION IN IDENTIFICATION OF LINEAR
MULTIVARIABLE MULTI-CONNECTED DISCRETE SYSTEMS
Abstract. The article deals with the problem of structural and parametric identification of a complex multivariable multi-input multi-output (MIMO) discrete system in state space model class. It is assumed that only the input and output coordinates of the system during certain time interval and range of measurement errors are known. The basis is the subspace (4SID) method, which, however, assumes that dimension of the system (state vector) is known, which is not always feasible in practice. Moreover, depending of the noise level, it is impossible to correctly identify a high-dimensional system. Therefore, it is proposed to use dimension as a regularizing parameter. Three methods for choosing of approximate model dimension are suggested depending on the length of the observation period and possibility of active experiment design. The proposed methods are verified on the example of identification problem of a commercial bank’s cognitive map in impulse process.
Keywords: structural and parametric identification, approximate model, subspace method (4SID), regularization, MIMO system, cognitive map, impulse process.
FULL TEXT
REFERENCES
- Ninness B., Goodwin G.C. Estimation of model quality. Automatica. 1995. Vol. 31, N 12. P. 1771–1797.
- Giarre L., Kacewicz B.Z., Milanese M. Model quality evaluation in set membership identification. Automatica. 1997. Vol. 33, N 6. P. 1133–1139.
- Kailath T. Linear systems. Englewood Cliffs, N.J.: Prentice Hall, 1980. 680 р.
- Ho B., Kalman R.E. Effective construction of linear state-variable models from input/output functions. Regelungstechnik. 1966. Vol. 14. P. 545–548.
- Verhaegen M., Dewilde P. Subspace model identification. Part 1: The output-error state space model identification class of algorithms. International Journal of Control. 1992. Vol. 56, N 5. P. 1187–1210.
- Van Overschee P., De Moor B. N4SID: Subspace algorithms for identification of combined deterministic-stochastic systems. Automatica. 1994. Vol. 30, N 1. P. 75–93.
- Verhaegen M. Identification of the deterministic part of MIMO state space models given in innovations form from input-output data. Automatica. 1994. Vol. 30, N 1. P. 61–74.
- Van Overschee P., De Moor B. Subspace identification for linear systems. Boston; London; Dordrecht: Kluwer Academic Publications, 1996. 254 p.
- Gubarev V.F. Rational approximation of systems with distributed parameters. Kibernetika i sistemnyj analiz. 2008. N 2. P. 99–116.
- Tikhonov A.N., Arsenin V.Ya. Methods for solving incorrect tasks. Moscow: Nauka, 1986. 286 p.
- Viberg M. Subspace-based method for the identification of linear time-invariant systems. Automatica. 1996. Vol. 31, N 12. P. 1835–1851.
- Ljung L. System identification Toolbox. User’s guide. Natick, MA: Mathworks, 2014. 886 p.
- Горелова Г.В., Захарова Е.Н., Радченко С.А. Исследование слабоструктурированных проблем социально-экономических систем. Когнитивный подход. Ростов-на-Дону: Изд-во РГУ, 2006. 332 p.
- Roberts F.C. Discrete mathematical models with applications to social, biological and environmental problems [Russian translation]. Moscow: Nauka, 1986. 496 p.
- Gubarev V.F., Romanenko V.D., Milyavsky Yu.L. Identification in cognitive maps in the mode of impulse processes with full information. Problemy upravleniya i informatiki. 2018. N 4. P. 30–43.
- Miliavskyi Yu.L. Identification in cognitive maps in impulse process mode with incomplete measurement of nodes coordinates. Кибернетика и вычислительная техника. 2019. N 1 (195). C. 49–63.
