Power Quality Enhancement In PV Based AC Microgrid Using ANN Controller

Abstract

Solar power is considered a very promising source for electric power generation. The increased penetration of PV based microgrid in distributed feeder system leads the power quality issues, especially under islanded condition. So there is need for efficient controlling scheme which can efficiently regulate the output of the system along with it improves the power quality. In this dissertation, artificial neural network is proposed as the inverter control system in PV based microgrid as an alternative to standard PI inverter control scheme. PV microgrid is connected at optimally position in 13 IEEE node feeder system. The controller model has nine numbers of input signals and generates twelve pulse output signals for inverter circuit. Artificial neural network is most suitable for nonlinear and complex system and its response is better than other controlling scheme. Due to this feature, ANN is proposed as inverter control scheme in system and it improves the performance and efficiency of the inverter and enhances the power quality. The proposed ANN trains feedforward network using Levenberg-Marquardt algorithm. The proposed system is verified through MATLAB simulink and the results are compared between the PI and neural network based controllers. The total harmonic distortion, voltage and phase angle variations are monitored and maintained within satisfactory range for the enhancement of microgrid power quality during grid-connected and islanded modes of operation.