Scaled Conjugate-Artificial Neural Network-Based novel framework for enhancing the power quality of Grid-Tied Microgrid systems

The exponential increase in the dependence on electrical power by rapid industrialization and increasing population has given rise to the need of supplying electric power efficiently and with good power quality (PQ). The evolution of renewable energy sources and their increased penetration into traditional grid systems has increased the complexity of electric power transmission and distribution. The significant increase in the use of complex converters, Flexible AC Transmission System devices and complex nonlinear loads in modern power systems have amplified the issues associated with the production and distribution of electricity having proper PQ. The proper functioning of the power system network requires healthy quality of power, maintained both at the load and source end. The Dynamic Voltage Restorer is a prospective D-FACTS (Distribution Flexible AC Transmission System) device that has been widely incorporated for addressing the PQ issues caused by irregularities in the distribution grid's voltage, current, or frequency. This research article attempts to enhance PQ indices of Photovoltaic, Fuel Cell, DVR and energy storage (battery) based Microgrid systems through a proposed Scaled Conjugate based Artificial Neural Network (SC-ANN) in the Matlab/Simulink architecture. The suggested technique's effectiveness is verified by introducing severe PQ faults such as sag and swell and numerous power system responses have been comprehensively studied by comparing them with traditional Fuzzy Logic Controller and Proportional Integral controllers. The results obtained validates the efficiency of the proposed controller over FLC and PI methods in maintaining the system power indices constant during the onset of PQ faults by reducing harmonics and oscillations thereby enhancing the overall system reliability, efficiency and stability paving its way for real-time implementation in the areas of MG system.