ANALYZING TEMPORAL AND SPATIAL VARIATIONS OF DIRECT NORMAL, DIFFUSE HORIZONTAL AND GLOBAL HORIZONTAL IRRADIANCES ESTIMATED FROM AN ARTIFICIAL NEURAL NETWORK BASED MODEL

摘要

An artificial neural network (ANN) ensemble approach has been successfully applied to estimate the direct normal irradiance (DNI) and diffuse horizontal irradiance (DHI) at the surface in the United Arab Emirates. Six thermal channels of the SEVIRI instrument, onboard Meteosat Second Generation Satellite were used to generate the model. Additional inputs are the solar zenith angle, latitude, longitude, solar time, day number and eccentricity correction. The global horizontal irradiance (GHI) is then calculated from DNI and DHI estimates. This study assesses the temporal variation of the estimated DNI, DHI and GHI through comparisons with ground measured values. For the three ground measurement stations available in the study area, the temporal assessments include data from heavy dusty, moderate dusty and clear days. The study also assesses the spatial variation through the visualization of DNI, DHI and GHI maps at specific scenes for the different weather conditions.