Controle preditivo direto pela potência aplicado aos conversores do gerador de indução duplamente alimentado

Abstract

Model-based predictive controls (MPCs) stand out for their simplicity, using the plant’s mathematical model as the basis for their control, and their high dynamic performance. In this work, the predictive control with finite control set (FCS-MPC) of the power is presented, to control, through the AC-DC-AC converter, also known as back-to-back, the stator power of the DFIG, the grid’s power factor, and the DC-link voltage. This control is based on the dynamic equations of the system to predict the behavior of the control variables in all possible switching states of a two-level power converter, which can be applied constantly, throughout the entire sampling time. In this way, the switching state that minimizes the error between the expected quantities and their references is selected. The behavior of the system with a DFIG of 0.56𝑘𝑊 is analyzed in a computational environment. PSIM ® software is used to obtain the behavior of the system variables and MATLAB ® in the graphical presentation of the results of the main quantities in steady-state and in the stator active power transient. These results show good dynamics in power control, low torque oscillation, and meet the requirements of the IEEE 512-2014 standard on total harmonic distortion (THD) in the stator current. To validate control performance, the results are compared with classic Direct Power Control (DPC) techniques. In addition, the impact of the reduction of active vectors is studied, seeking a reduction in the computational burden of the control and, an observer that mitigates the errors caused by the parametric variation of the generator. With the reduction of active vectors, there is a reduction in the computational cost with a low impact on the power control. Experimental results are also presented and discussed, to verify the functioning of the studied controls.