Stopping rules formation and faults detection in parametric identification problems


Hacızade C.

Proceedings of the Institution of Mechanical Engineers. Part I: Journal of Systems and Control Engineering, cilt.215, sa.4, ss.357-363, 2001 (SCI-Expanded) identifier

Özet

An approach to the generation of stopping rules in parametric identification problems is proposed on the basis of the computation of a statistic of the difference between two successive estimates. This statistic is also used for fault detection in the Kalman filter. The developed decision rules are applied to a linear system identification problem. Experimental results are presented to demonstrate the performance of the proposed algorithms.