Contribution to the Neural network speed estimator for sensor-less fuzzy direct control of torque application using double stars induction machine

The main objective of this paper is to study of adaptive speed estimator for a double start induction machine using an artificial neural network to estimate the speed with a fuzzy direct control of torque for the converter switches. The estimation algorithm uses the current& voltage stator values combined with an intelligent adaptive mechanism (MRAS) based on an artificial neural network (ANN) to estimate rotor speed, also a simple Proportional-Integrator (PI) used as speed controller. Thus hysteresis comparators used on the classical method of direct control of torque has been replaced by fuzzy blocs. As results we achieved can be summarised as follows: 1-amelioration the responding time of the system 2-Minimization of the torque ripples. 3-Minimization of the current total harmonic distortion.