Solar power forecasting of a residential location as part of a smart grid structure

This paper presents the use of an artificial neural network for classification on a residence house that uses local air temperature and solar insulation predictions to identify patterns at the desired location, in order to obtain a stochastic distribution of the daily solar profile. This is a first step on the further creation of a short-term operation model that allows determining the technical and economic impact of stationary/mobile batteries of electric vehicles in presence of microrenewables. This short-term operation model will be in the day-ahead perfect market operation (unit commitment) where specific changes are made to consider stationary and mobile operation.