The PI controller research of UPQC in micro-grid based on RBF neural network

Aiming at the micro-grid deteriorating the distributed network power quality for its features like voltage fluctuation and intermittent, unified power quality conditioner (UPQC) is the most effective device for power quality treatment. The voltage sag detection of UPQC based on PI with RBF (Radial Basis Function) neural network controller is put forward in the paper. The UPQC topology which includes photovoltaic array is constituted. The PI control strategy based on RBF neural network is studied, and algorithm steps are summarized. After that the compensation instruction controlled by presented PI controller is used to modulate pulse width, and the pulses drive universal bridge to compensate voltage. The simulations show that the proposed PI controller has a better voltage compensation effect than the traditional PI controller through simulation comparison. It can detect voltage sag quickly and improve the micro-grid power quality. The new control strategy has strong anti-jamming capacity and robustness. It provides a new control strategy for UPQC treating power quality of micro-grid. © 2016, UK Simulation Society. All rights reserved.