Analysis and validation of 24 hours ahead neural network forecasting of photovoltaic output power

7 In this paper an artificial neural network for photovoltaic plant energy fore8 casting is proposed and analyzed in term of its sensitivity with respect to the 9 input data sets. 10 Furthermore, the accuracy of the method has been studied as a function 11 of the training data sets and error definitions. The analysis is based on exper12 imental activities carried out on a real photovoltaic power plant accompanied 13 by clear sky model. 14 In particular, this paper deals with the hourly energy prediction for all 15 the daylight hours of the following day, based on 48 hours ahead weather 16 forecast. This is very important due to the predictive features requested 17 by smart grid application: renewable energy sources planning, in particular 18 storage system sizing, and market of energy. 19