Estimation of Global Solar Radiation in India Using Artificial Neural Network

In this study, global solar radiation was predicted using artificial neural networks. Gradient descent back propagation with adaptive learning rate was used for training the artificial neural network. In order to train and test the neural network, meteorological day average data like maximum ambient temperature, minimum ambient temperature and minimum relative humidity values were used for predicting global solar radiation in future time domain using artificial neural network method. The measured data was randomly selected for training and testing the neural networks. Obtained results show that using the minimum ambient temperature and day of the year outperforms the other cases with absolute mean percentage error of 6.65% and mean squared error of 0.008. This neural network, therefore, can be used for estimating global solar radiation for locations where only ambient temperature data are available.