×

Parametric identification of large scale discrete time systems. (English) Zbl 0677.93013

Summary: This paper deals with the parametric identification of a large-scale discrete-time system decomposed into interconnected subsystems. We propose some recursive identification algorithms based on the least squares method. These algorithms are analysed in the deterministic and stochastic environment. We have established necessary and sufficient convergence conditions for the deterministic algorithms from the hyperstability and positivity concepts. Conditions for convergence of the stochastic algorithms are presented by using the ordinary differential equation method. The validity and effectiveness of the theoretical results are tested by numerical simulations.

MSC:

93B30 System identification
93C55 Discrete-time control/observation systems
93E15 Stochastic stability in control theory
93A15 Large-scale systems
PDFBibTeX XMLCite