Modelo para la predicción de la radiación solar a partir de redes neuronales artificiales

Abstract

The estimation of the climatic change that occurs in a particular place has always been an important factor for doing outdoor activities. Different methods have been developed to predict future values of climatic variables which can provide information to people and companies, allowing them to organize their activities. The solar radiation is one of these climatic variables and it has become fundamental in multiple circumstances, especially in photovoltaic centers. The project’s goal is to give an approximated value of the solar radiation that impacts the solar vehicles of the World Solar Challenge 2013 in five locations close to the road where the competition occurs; for this to be carried out, a model based in neural networks is analyzed and developed. It also seeks to find the value of the solar radiation at intermediate points of the five mentioned before using space radial functions.