Modelo predictivo inteligente para la generación de energía eléctrica en un sistema fotovoltaico

Abstract

Solar energy is currently the most resourceful renewable source worldwide in which its obtainment, unlike many others currently used, does not mean any harm to the environment. To manage satisfactorily the electricity demand of photovoltaic systems a careful planning is required, nonetheless, to set successfully the solar behavior and even more to predict the amount of incident solar radiation is a completely stochastic process due the influence of meteorological variables. To overcome this problem, this study monitors diverse atmospheric variables by using a meteorological station located in the University of Sonora in Hermosillo city, Mexico; and registers real-time measure values every 5 minutes for 6 months. From the data obtained a computational model applying an Adaptive Neuro Fuzzy Inference System (ANFIS) as intelligent technique is presented in order to forecast the electrical power from a photovoltaic system installed in the same site. It is vital to understand the influence of meteorological variables on energy consumption in which a better understanding of it can contribute to a more useful strategy in meeting energy efficiency goal for Mexico.