Speed estimator in closed-loop scalar control using neural networks

Abstract

This work proposes an artificial neural network approach to estimate the induction motor speed applied in a closed-loop scalar control. The induction motor speed is the important quantity in an industrial process. Thus, when the load coupled to the axis needs speed control, some of the drive and control strategies are based on the estimated axis speed of the motor. This paper proposes an alternative methodology for estimating the speed of a three phase induction motor driven by a voltage source inverter, using space vector modulation under the scalar control strategy and based on artificial neural networks. Experimental results are presented to validate the performance of the proposed method under motor load torque and speed reference set point variations.