Microprocessor implementation of field-oriented control of induction motors using ANN observers

Abstract

This paper presents a microprocessor based intelligent controller implemented, in a single processor, for field-oriented induction motor control taking the non-linear parameter variation into account. Previous investigations [2,10] on this subject have neglected the effect of saturation in the air gap flux and hence the corresponding non-linear parameter variation of the induction motor. As the phase angle of rotor magnetizing current (or m.m.f. vector) in a standard induction motor cannot be measured by direct means an observer is generally needed in field-oriented control of induction motors. Two types of observers (based on the linear and non-linear model of the machine) have been used in field-oriented induction motor control schemes [1,2,10]. The reduced-order linear model of observer [10] is easy to implement in real time, but it does not give an accurate estimation of the rotor m.m.f. vector angle, β, since the induction motor can operate in the region of saturation. The non-linear observer model which incorporates this effect of magnetic saturation of the induction motor cannot be practically implemented by using normal methods as it takes too long a time to estimate the angle β. The implementation of the real-time intelligent controller in this project is based on Artificial Neural Networks (ANN) which take into account the effect of saturation and estimate the angle β in a few microseconds which is well within the real time deadline.