Controle Data-Driven: projeto de controlador utilizando o método VRFT aplicado ao controle de velocidade em motor de indução

Abstract

This work presents the implementation of the Virtual Reference Feedback Tuning (VRFT) method, a data-driven approach for tuning a discrete PI controller, applied to a three-phase induction motor. The study is conducted through experimental tests performed in a laboratory bench, allowing a practical evaluation of the controller’s performance against different first-order reference models. The tests consider step and sinusoidal disturbances, enabling a comparative analysis of the integral of absolute error (IAE) and total variation of control (TV). The results demonstrate that the choice of the reference model directly influences the control robustness and disturbance rejection, highlighting the need for a trade-off between stability, response time, and control effort in controller tuning.