ANN-based modelling and estimation of daily global solar radiation data: A case study

In this paper, an artificial neural network (ANN) models for estimating and modelling of daily global solar radiation have been developed. The data used in this work are the global irradiation HG, diffuse irradiation HD, air temperature T and relative humidity Hu. These data are available from 1998 to 2002 at the National Renewable Energy Laboratory (NREL) website. We have developed six ANN-models by using different combination as inputs: the air temperature, relative humidity, sunshine duration and the day of year. For each model, the output is the daily global solar radiation. Firstly, a set of 4 × 365 points (4 years) has been used for training each networks, while a set of 365 points (1 year) has been used for testing and validating the ANN-models. It was found that the model using sunshine duration and air temperature as inputs, gives good accurate results since the correlation coefficient is 97.65%. A comparative study between developed ANN-models and conventional regression models is presented in this study.