ANFIS-based hysteresis comparators with intelligent dual observer and speed controller of a direct torque control

Abstract

This article presents the adaptive-network-based fuzzy inference system (ANFIS)-based direct torque control (DTC) for induction motor (IM). DTC is distinguished by merging a simple structure with a good dynamic behavior. Despite these cited advantages, some disadvantages are also present. For this aim, which consists of reducing the ripples in electromagnetic torque, flux and current; and to improve the IM response characteristics, the conventional hysteresis comparators of the torque, flux and the proportional integral (PI) speed controller are replaced by others based on ANFIS technique. Furthermore, an intelligent dual observer is implemented to achieve sensorless control; merging a Luenberger observer (LO) based on ANFIS to insure the adaptation mechanism in order to estimate the rotor speed, and a Kalman filter (KF) to insure the flux components estimation. The proposed sensorless ANFIS-DTC shows a robust performance; reduced ripples, decreased overshoots, short time of rising and settling, and high resistance to perturbations.