Control directo del par de un motor de inducción utilizando redes neuronales

Abstract

This paper deals with speed control for a squirrelcage induction motor by using direct torque control (DTC) through the Simulink® professional simulation program for later implementation. The control maintains the electromagnetic flux and torque of the motor constant, for which it estimates the flux and the torque measuring the voltages and currents in the stator. In addition it selects the suitable voltage vector according to the angle of the electromagnetic flux, and increase or decrease the torque or flux. In this work the selection of the appropriate voltage was made with the use of an artificial neural network, in order to determine if this contributes to decrease the torque and flux ripple, because the direct control of the torque is characterized by having a dynamic fast response but a high torque ripple.