Frequency analysis of stator currents of an induction motor controlled by direct torque control associated with a fuzzy flux estimator

Abstract

Introduction. The best way to control the torque of an induction motor is conventional direct torque control (DTC); this control method is the most used approach in the industrial sector due to its many advantages. Its main advantages are its simplicity and its exclusive dependence on the stator resistance of the induction motor. However, the use of hysteresis comparators reduces its effectiveness, causing more torque ripple. Additionally, this results in variable operating frequency and limited frequency sampling, resulting in pseudo-random overshoot of the hysteresis band. Purpose. For these reasons, this article presents a new study aimed at confirming its shortcomings and improving the effectiveness of the control. Novelty. We propose to use fuzzy logic methods to estimate the two components of the stator flux. Methods. In traditional DTC the flux components are estimated from an equation relating the stator resistance to the stator voltage and current. In the proposed method, only stator currents and voltages are used for this evaluation, which eliminates the dependence of DTC on stator resistance. The aim of this proposal is to make DTC robust to parametric changes. Results. General harmonic distortions, rotational speed of the induction motor, electromagnetic moment, magnetic flux and stator currents are analyzed. Practical value. With this proposed technique, validated in Simulink/MATLAB, several improvements in motor behavior and control are endorsed: torque fluctuations are reduced, overshoot is completely eliminated, and total harmonic distortion is significantly reduced by 48.31 % for stator currents. This study also confirmed the robustness of DTC to changes in stator resistance.