Stator resistance tuning based on a neural network in an indirect rotor field oriented control system of an induction motor

Abstract

This paper presents an ANN-based (artificial neural network-based) method of stator resistance tuning in an IRFO (indirect rotor field oriented) control system of an induction motor. This method is based on the conventional two-layer ANN in which the rotor time constant is not a constant parameter and is identified using a model reference adaptive system (MRAS) - based procedure. During the training, rotor speed estimation of the induction motor is enabled. The difference between the actual and the estimated rotor speed is used as a signal for manual stator resistance tuning. Computer simulations and experimental results show the effectiveness of the described approach in a low rotor speed region. © 2008 Springer-Verlag Berlin Heidelberg.