Best Combinations of Inputs for ANN-Based Solar Radiation Forecasting in Morocco

Abstract

In this paper, for forecasting, with the high possible accuracy, the solar radiation intensity, an approach for identifying the optimum set of input data from large sets of input parameters is elaborated, assessed, and proposed. This approach belongs to techniques of Input Variable Selection (IVS) and was established using artificial neural networks (ANNs) as an effective means for numerically approximating the computed objective function: Monthly Average Global Solar Radiation (MAGSR). Based on the approach suggested, we could determine the best combinations of inputs that can reveal great possible correlation and approximation with the output considered parameter. Recorded data from 35 stations of different climatic zones were used for both training and testing purposes. Several new linear formulas between the MAGSR and other climatological and meteorological parameters were developed and evaluated. Based on these relationships, we can forecast other climatological and meteorological parameters like temperature, wind speed, and humidity. The statistical analysis was done, and the best performance of the proposed approach has been well checked and duly validated.