Prediction and performance assessment of global solar radiation in Indian cities: A comparison of satellite and surface measured data

The design and performance assessment of solar photovoltaic/thermal system requires the knowledge of global solar radiation. Generalised models to assess the monthly average global solar radiation of twelve locations in India were established using sunshine hours available for 14 years (2000-2014) obtained from the World Radiation Data Centre (WRDC), Russia. Statistical analysis is carried out to assess the accurate model to estimate the global solar radiation which includes mean bias error, root mean square error, mean absolute percentage error, the coefficient of determination and uncertainty at 95%. Also, the assessment of the global radiation values obtained from satellite data by the National Renewable Energy Laboratory (NREL) is performed. Further, Global Performance Indicator (GPI) is introduced to find the ranking of models from the overall performance of the investigated models by scaling statistical indicators. The comparison of the NREL estimated data and WRDC data shows that the satellite data overestimates the surface measured data between 10 and 15% for seven locations, whereas, for the remaining four stations, it is between 7 and 9.5%. (C) 2019 Elsevier Ltd. All rights reserved.