Controle preditivo de corrente com robustez a dependência paramétrica e abordagem dois vetores

Abstract

Finite Control Set Model Predictive Current Control (FCS-MPCC) provides a robust method for controlling electric machines, such as induction motors, by selecting appropriate voltage vectors to be applied through an inverter at each time instant using an algorithm. However, in the classical approach, its effectiveness relies on accurate motor parameters, which are not always available, thereby affecting both dynamic and steady-state performance. To enhance this behavior, a hybrid scheme has been developed, combining the Two-Vector Model Predictive Current Control (MPCC2V) approach and the Model-Free Predictive Current Control (MPCCMF) through an Extended Linear State Observer (LESO). Real-time input/output data and estimated parameters are utilized, leading to the proposed method named Two-Vector Model-Free Predictive Current Control (MPCC2VMF). This approach reduces sensitivity to parameter uncertainties and minimizes current and torque ripples by employing a modulator. Finally, computational simulations based on a 1 HP induction motor demonstrate that the MPCC2VMF method successfully leverages the advantages of both techniques, ensuring stable performance even in the presence of parameter inaccuracies.