An efficient observer model for field oriented induction motor control

An observer is needed in field-oriented induction motor control as the phase angle of rotor magnetising current or m.m.f. vector in a standard induction motor cannot be measured by direct means. Two types of observers (linear and non-linear) are used in field-oriented induction motor control schemes as can be found in literature. The reduced-order linear model of the observer is easy to implement in real time, but it does not give an accurate estimation of m.m.f. vector angle, /spl beta/, since the induction motor operates in the region of saturation. The non-linear model which incorporates this effect of magnetic saturation of the induction motor cannot be practically implemented as it takes too long a time to estimate the angle /spl beta/. This paper presents a real-time computationally-efficient observer based on artificial neural networks (ANN) which takes into account the effect of saturation and estimates the angle /spl beta/ in a few microseconds which is well within the real time deadline.