Improvement of DC Power Response Using Artificial Neural Network in an Off-Grid PV System Linked to a Constant Power Load

With the growth of renewable energies, solar electricity is applied to power various loads like electric motors and illuminants, researchers conduct extensive studies on achieving electricity production with high power quality. The output power from the photovoltaic system undergoes considerable variability, necessitating the use of maximum power point tracking (MPPT) algorithms. In this article, the traditional P&O controller and the intelligent fuzzy logic (FL) controller are initially examined. Subsequently, a fast and reliable artificial neural network (FR-ANN) controller is formulated, designed to perform effectively under conditions of irradiation and variable temperature by assimilating favorable attributes from other controllers through learning. The photovoltaic system is simulated in the MATLAB/Simulink environment and provides power to a constant power load sensitive to voltage drop. Through the application of the FRANN controller, an improvement in system stability is observed in the event of a voltage drop.