Artificial Intelligence-Assisted SVPWM for Enhancing Power Quality in Solar Photovoltaic Systems

Abstract

The goal of this research is to improve the efficiency and dependability of grid-connected solar photovoltaic (PV) systems by utilizing Artificial Neural Networks (ANN) and Space Vector Pulse Width Modulation (SVPWM). The main issues with power quality were reducing total harmonic distortion (THD) and enhancing voltage control. In the field of solar photovoltaic (PV) systems, these issues are crucial as power electronics converters and the amount of incoming solar radiation are both subject to fluctuation. An artificial neural network (ANN) technique uses the high-quality voltage waveform and real-time power regulation of the SVPWM inverter to dynamically adjust the inverter's switching patterns. We were able to conclude that the suggested method worked better than the conventional SVPWM processes after conducting simulations in a number of work situations. One noteworthy discovery was the average overall harmonic distortion, which is now being lowered to less than 5% in order to comply with IEEE 519 standards. The efficiency of the ANN-SVPWM system was 96% higher than that of the conventional one, and it maintained a steady voltage even when the sun's irradiation levels changed. Furthermore, this function demonstrated that grid-connected PV systems may respond more quickly to changes in the sun's conditions in real time, perhaps making them a more effective option. In the biological sciences, researchers discovered that combining artificial neural networks (ANN) with systematic vector pulse width modulation (SVPWM) enhanced the dependability, efficiency, and power quality of solar photovoltaic (PV) systems.