The use of a sky camera for solar radiation estimation based on digital image processing

• Using a sky camera, the solar resource has been estimated for one minute periods.
• The sky images have been processed to estimate the solar radiation at pixel level.
• The three radiation components have been estimated under all sky conditions.
• Results have been presented for cloudless, partially-cloudy and overcast conditions.
• For beam and global radiation, the nRMSE value is of about 11% under overcast skies.

The necessary search for a more sustainable global future means using renewable energy sources to generate pollutant-free electricity. CSP (Concentrated solar power) and PV (photovoltaic) plants are the systems most in demand for electricity production using solar radiation as the energy source. The main factors affecting final electricity generation in these plants are, among others, atmospheric conditions; therefore, knowing whether there will be any change in the solar radiation hitting the plant's solar field is of fundamental importance to CSP and PV plant operators in adapting the plant's operation mode to these fluctuations. Consequently, the most useful technology must involve the study of atmospheric conditions. This is the case for sky cameras, an emerging technology that allows one to gather sky information with optimal spatial and temporal resolution. Hence, in this work, a solar radiation estimation using sky camera images is presented for all sky conditions, where beam, diffuse and global solar radiation components are estimated in real-time as a novel way to evaluate the solar resource from a terrestrial viewpoint.