Machine Learning Technics for the prediction of extreme overirradiance events
Machine Learning, Artificial Intelligence, Solar Energy, Overirradiance, LSTM
Due to the advent of climate change, there was a rapid development in new forms of energy generation, among which photovoltaic generation stands out for the abundance of generation resources and for its low maintenance and installation cost. However, due to the strong correlation between meteorological variables and solar generation, climatic phenomena that were previously little explored become more relevant for the prevention of economic losses in large-scale generation units.
In this context, this work seeks to obtain a greater understanding of extreme overradiance events from the perspective of computational intelligence, for this purpose, from a set of meteorological data with a temporal resolution of 1 second, a set of machine learning algorithms will be tested with the objective of predicting the occurrence of this phenomenon.