Short-term forecasting of grid-connected solar photovoltaic generation

The project aims on the enhancement of the capabilities of the partners to use meteorological forecasts for the improvement of the control of renewable energy systems and thus to ease the integration of these technologies to the energy supply schemes of their countries. Building on the experiences of the partners in the field of renewable energy systems, meteorology and metrology in solar and wind energy applications, a comprehensive scheme for an optimized use of meteorological forecast information for system control and thus better grid and system integration should be developed. This should comprise on one hand the combined schemes for the improvement of the numerical weather forecasts by post-processing with the inclusion of space and time information on historic forecast errors, and on the other hand the remaining forecast errors are to be characterized by their distribution function in order to obtain reliable confidence intervals. This information should form the input to schemes for the optimal probabilistic control of the renewable power systems. This is an international cooperation project between Universidade Federal de Santa Catarina (Prof. Ricardo Rüther), University of California - San Diego (Prof. Carlos F. M. Coimbra), Solar Energy Research Institute of Singapore - SERIS / National University of Singapore - NUS (Prof. Thomas Reindl and Andre Nobre), Instituto Nacional de Pesquisas Espaciais - INPE (Prof. Enio B. Pereira and Dr. Fernando R. Martins), and University of Agder / Norway (Prof. Hans-Georg Beyer).