Genetic algorithm applied to the specialization of neural networks for the forecast of wind and solar generation

Abstract

This work presents the training of a Neural Network (NN) using Genetic Algorithm (GA) for medium-term wind and solar power generation forecasts. Wind speed, solar radiation, and other meteorological variables forecasts are provided by a meteorological service and are used as the NN’s inputs. The NN’s output is the generated power forecast. A comparison between the previously used mdel and the new one s shown and future improvements are discussed