Analysis of Speed Estimation Techniques for Asynchronous Machines

L. Mouzinho and J.V. da Fonseca Neto (Brazil)


Induction Machine, estimation, neural networks, indirectmeasurement, sensorless.


This paper presents a comparative analysis of four tech niques for estimation of speed in Asynchronous Machines: Rotor flux, electromotive force (emf), model reference adaptive system and neural networks are the compared techniques via computational simulations. An analysis is performed to verify the time expense of the estimation, the sensitivity in the variation of the parameters and difficulty in frequencies next to zero. A major concern is to obtain an estimator design that can be used in real time implementa tions. Therefore, this research seeks for fast and accurate estimators, considering its numerical calculations and the machine parameters uncertainties that gives a rotor speed approximation as close as possible of the real world appli cations.

Important Links:

Go Back