Assessment Of Daily Global Solar Radiation Using Radial Basis Function Techniques

In this study, many experiments were carried out to assess the influence of the input parameters on performance of the radial basis function network (RBF), which is one of the architectures of neural networks. To estimate the daily global solar radiation on horizontal surface, we have developed some models by using seven combinations of 12 meteorological and geographical input parameters, collected from a radiometric station installed at Ghardaia site (southern of Algeria). For the selection of best combination which provides a good accuracy, six statistical formulas (or statistical indicators) have been evaluated, such as the root mean square errors, mean absolute errors, correlation coefficient, and determination coefficient. We noted that RBF techniques have the best performance, except when the sunshine duration parameter is not included into the input variables. Accordingly, the maximums of determination coefficient and correlation coefficient are equal to 98.20 and 99.11% respectively. On the other hand, some empirical models were developed in order to compare their performance with those of multilayer perceptron neural networks. Results showed that the RBF networks techniques have best performance compared to the empirical models.