Adaptive neuro-fuzzy inference system into induction motor: Estimation

This paper presents the application of an adaptive neuro-fuzzy inference system (ANFIS) for an induction motor for speed estimation. Due to the drawbacks of the mechanical sensors, ANFIS (neuro-fuzzy inference adaptive system) speed observer is developed and it is based on artificial intelligence technique combining the concepts of fuzzy inference systems and neuron networks. The ANFIS rotor speed estimator depends only on measurable stator quantities (voltages and currents) that are easily accessible, hence the easy implementation in practice and thus reduces the cost since there is no need to the speed sensor. In addition, this work deals also with the vector controlled induction motor using stator field orientation (SFO). It is well known that the vector control strategy is based on the simultaneous determination of the magnitude and argument of the flux vector. This control method gives an effective solution that provides decoupling between the flux and torque of an induction motor, hence overcoming the complex control obstacle of this type of machines. Simulations to evaluate the performance of the estimator considering the vector drive system were done from the Matlab/Simulink software.