A flux estimator for field oriented control of an induction motor using an artificial neural network

Abstract

Field oriented control (FOC), sometimes called vector control, is used in inverter-fed induction motor drives to obtain high performance speed response. For field oriented control it is necessary to know the instantaneous magnitude and position of the rotor flux. The magnitude and position of the rotor flux is approximated based on flux measurements in the direct FOC scheme and estimated in the indirect FOC scheme. In this paper a novel flux estimator, the neural network flux estimator, is presented. The neural network is able to estimate accurately the rotor flux magnitude or position (maximum absolute error is less than 0,03 p.u.) for line-start operation of an induction motor. Its ability to estimate flux response that lies outside of the neural network training data set is another one of its strengths. The authors' preliminary work indicates that neural network flux estimation may be a feasible alternative to other flux estimation methods. >