Solar radiation forecasting based on meteorological data using artificial neural networks

The main objective is to predict daily global solar radiation (GSR) in future time domain based on measured air temperature, relative humidity and sunshine hours values between 2002 and 2006 for Dezful city in Iran using artificial neural network method. The estimations of GSR were made using three combinations of data sets: (I) length of day, daily mean air temperature and relative humidity as inputs and GSR as output, (II) length of day, daily mean air temperature and sunshine hours as inputs and GSR as output, (III) length of day, daily mean air temperature, relative humidity and sunshine hours as inputs and GSR as output. The measured data between 2002 and 2005 were used for training the neural networks while 235 days´ data from 2006 as testing data. The testing data were not used in training the neural networks. Obtained results show that neural networks are well capable of estimating GSR from simple and available meteorological data. This can be used for estimating GSR for locations where only simple meteorological data are available.