IDENTIFICATION BY HYBRID ALGORITHM APPLICATION TO THE IDENTIFICATION OF DYNAMICS PARAMATERS OF SYNCHRONOUS MACHINE

This work treats the identification of the dynamic parameters of a synchronous machine using the test by symmetrical three-phased short circuit. The fitting of these parameters is carried out through the minimization of the quadratic errors between the short circuit currents measured and calculated.. The approach of hybridization used in this work is the combination of the genetic algorithm (AG) and the algorithm of Levenberg-Marquardt (LM). The results of the optimization are validated first by comparing the parameters obtained by the approach adopted with those estimated with the graphic method of the IEEE standard then by the test of "voltage of reappearance".