Scaled Conjugate Gradient Function Fitting Based Artificial Neural Network to Mitigate Power Quality Issues in an AC Micro Grid

Abstract

The scarcity of conventional energy sources leads to the exponentially use of renewable energy sources to meet the huge power demand. However, using renewable sources is challenging because of grid integration problems, energy loss, and emission of polluting agents. Micro Grid (MG) plays a vital role by replacing the traditional depleting resources with renewable and non-traditional resources. The alternate energy sources like solar, wind and fuel cell are environmentally friendly and used as Distributed Generators (DG) in the MG concept, but due to the nonlinear characteristics of the sources and their environmental dependency, the Power Quality (PQ) issues arise during MG operation. Thus, it is essential to design an intelligent controller to maintain the PQ issues within the specified values. In this paper, a Scaled Conjugate Gradient Function Fitting based Artificial Neural Network (SCGFANN) with centralized distribution static compensator controller has been designed for the small and large MG and is simulated in MATLAB/Simulink environment. The effectiveness of the proposed controller has been investigated in various conditions such as voltage sag/swell, unbalance, disturbances, and frequency distortion in the case of a simple MG. The analysis of the switching delay in the controller action has been performed to scrutinize the significance of the suggested controller in the case of a complex practical MG. It can be observed from the simulation results that the proposed controller outperforms the conventional Fuzzy-PI controller in terms of better stability and enhanced PQ mitigation, thus justifying its real-time implementation.