Direct Power Neuro-Fuzzy Controller Scheme of Three-Phase PWM Rectifiers for Power Quality Improvement

Abstract This paper investigates the design of an improvement scheme of direct power control (DPC) based on adaptive neuro-fuzzy inference system (ANFIS) for mitigation of power quality problems. The proposed control technique is performed using two ANFIS blocks. The first ANFIS block uses the grid voltage vector position, the errors of the instantaneous active and reactive powers. The second ANFIS block utilizes the dc link voltage error and its change. The purpose of the proposed control algorithm is to regulate perfectly dc link voltage, to control directly the instantaneous powers, near-sinusoidal source current waveforms and unity power factor (UPF) operation. The regulation method has been developed and simulated using MATLAB/Simulink environment. The numerical simulation results confirm the validity of the proposed control technique for three-phase PWM rectifiers and have proved excellent robustness and performance.