Neural network and fuzzy logic in a speed close loop for DTC induction motors

Abstract

Direct Torque Control (DTC) is known to produce quick and robust response in AC drives. However, during steady state, torque, flux and current ripple occur. An improvement of the electric drive can be obtained using a DTC scheme based on the Space Vector Modulation (SVM) which reduces the torque and flux ripple. The proposed control scheme considers the rotor resistance variation. This paper also discusses the application of Type-2 Fuzzy speed Control under uncertain stimuli and an Artificial Neural Network (ANN) as a speed estimator. The capability and precision of this scheme as a speed controller and estimator are verified by different conditions and it is concluded that the proposed control scheme produces good results.