Modeling the forecasted power of a photovoltaic generator using numerical weather prediction and radiative transfer models coupled with a behavioral electrical model

Abstract The intermittency of the solar radiation is the big challenger for the management of electrical SMART GRID network. So the forecasting of the solar radiation remains the crucial objective to anticipate the injection of the electrical power produced by a photovoltaic generator. In this study, we use the Numerical Weather Prediction model WRF coupled to a radiative transfer model libRadtran to forecast the global, direct and diffuse solar radiations. The NMM core of WRF model was used to predict the vertical profiles of the atmospheric meteorological parameters such as liquid and ice water concentration, cloud cover, relative humidity, temperature, etc. These data are used as input in libRadtran to calculate the hourly solar radiations components (Direct, Diffuse and Global) in real meteorological situations. The radiation values are obtained on regular geographical grid points. A behavioral electrical power model is used to compute the electrical power produced by a photovoltaic generator located at the Center of Development of Renewable Energy in the city of Algiers in Algeria. The comparative study shows well that the results are very encouraging.