GALDINO, J. F.; http://lattes.cnpq.br/0746801791799877; GALDINO, Josenildo Ferreira.
Résumé:
In the classical approach to system identification problem of identifying the system is
the assessment of parameters of a given model, using an algorithm that uses data input
and output of the process. In contrast, system identification with orthonormal basis functions
allows incorporating a partial knowledge about the dynamics of the system which
reduces the order and, consequently, fewer parameters are necessary in the. representation
of the model.
In this thesis we address the problem of identifying base systems using Laguerre functions.
It proposed a new approach for identifying systems with, bases Laguerre using
the relay experiment to obtain an initial estimate of the pole of Laguerre. The use of
experiment relay allows to obtain a partial knowledge of the system being identified. In
addition, several experiments of some processes first and second order were simulated with
Matlab.