Unified power quality conditioner for improving power quality using ANN with hysteresis control

This paper presents a new control design of unified power quality conditioner (UPQC) for harmonic compensation in a power distribution system. The topology of the UPQC is based on two 3-phase voltage source inverters (VSIs) which share two DC link capacitor with midpoint grounded. The extraction circuit using an artificial neural network (ANN) controller with improved weights updating algorithm is proposed. Linear quadratic regulator (LQR) control technique is used to coordinate the operation of the series and shunt VSIs of the UPQC. LQR coordination ensures that the UPQC operates satisfactorily without depleting the limited energy of the DC link capacitors. Hysteresis control is used to generate accurate switching signals for the two VSIs. The simulation results show that the UPQC with neural network extraction circuit and LQR-coordinated hysteresis control can significantly reduce harmonic distortion in the supply voltage and load current and can adapt to variations in system operating conditions.