Experimental Identification of the Mechanical Parameters of an Induction Motor Drive

In order to obtain fast dynamic response performance of an induction motor drive, the identification of mechanical parameters such as the drive inertia and the coefficients of friction, with a good accuracy, is highly desirable. They are essential for the design of the high-performance induction motor drive speed, as well as position controllers and speed observers, since a drive response is influenced not only by load disturbances but also by these mechanical parameters. Moreover, they are of great importance for the accurate dynamic modeling and simulation of various high-performance induction motor control strategies. In this paper an experimental off-line method for the mechanical parameters identification is presented. The method uses speed-time curve, obtained during the retardation test on the drive, with an appropriate mechanical losses model of the drive, and the mean squared error performance function based on a genetic algorithm (GA) approach, to obtain unknown mechanical parameters of the tested drive. The proposed method is verified by experiments.