Deep Learning for Wind Speed Forecasting in Northeastern Region of Brazil

Deep Learning is one of the latest approaches in the field of artificial neural networks. Since they were first proposed in mid-2006, Deep Learning models have obtained state-of-art results in some problems with classification and pattern recognition. However, such models have been little used in time series forecasting. This work aims to investigate the use of some of these architectures in this kind of problem, specifically in predicting the hourly average speed of winds in the Northeastern region of Brazil. The results showed that Deep Learning offers a good alternative for performing this task, overcoming some results of previous works.