Enhancement of the Direct Torque Control by using Artificial Neuron Network for a Doubly Fed Induction Motor

Abstract

Direct Torque Control (DTC) is the most popular strategy used in the industrial sector, because of its various advantages, however, the torque ripples makes it less efficient, due to the use of the hysteresis comparators, leading to variable frequency operation and on the other hand, the finite frequency sampling results in a pseudo-random overshoot of the hysteresis band, Thus, operation at low speed and in particular, with variations in motor resistances, affects the behavior of the machine, in this reason, this article presents a new study to promote its drawbacks to increase the control performances. A new intelligent direct torque control applied to a Doubly Fed Induction Motor (DFIM) by two Vector Source Inverters (VSIs) based on an Artificial Neuron Network (ANN) who will replace the speed controller, switching tables, and hysteresis comparators, with this special technique simulated in Matlab/Simulink, approved several improvements on motor and control behaviors so as, the torque ripples has been improved by 55.82 %, the overshoot is absolutely removed and increasing important values of total harmonic distortion (THD) by 3.26 % and 3.31 % for stator and rotor currents respectively.