Abstract
This paper presents an original hysteresis PI based neural controller for speed control of an indirect rotor flux oriented controlled (IRFOC) induction motor drive. An original hysteresis PI controller is first proposed. Its simulated performances in speed control and rotor resistance estimation are then compared to those of a classical PI controller. The proposed hysteresis PI controller provides better dynamic performances than the classical PI controller but it has one drawback. The hysteresis PI cannot deal with down step speed tracking below a certain limit. The artificial neural network generalization capacity is then used to deal with this drawback. The simulated input-output non linear relationship of the proposed controller during startup and load disturbance rejection is learned off-line using a feed-forward linear network with one hidden layer. The simulation of the neural network controlled system shows promising results. The motor reaches the reference speed rapidly and without overshoot, step commands are tracked with almost zero steady state error and no overshoot, load disturbances are rapidly rejected and variations of some of the motor parameters are fairly well dealt with.