Forecast of ensemble power production by grid-connected PV systems

Abstract

The contribution of power production by PV systems to the electricitysupply is constantly increasing. Anefficient use of the fluctuatingsolar power production will highly benefit from forecast informationon the expected powerproduction. This forecast information is necessaryfor the management of the electricity grids and for solar energytrading. This paper will present and evaluate an approach to forecastregional PV power production. The forecast quality was investigatedfor single systems and for ensembles of distributed PV systems. Dueto spatial averaging effects the forecast for an ensemble of distributedsystems shows higher quality than the forecast for single systems.Forecast errors are reduced to an RMSE of 0.05 Wh/Wp for an ensembleof the size of Germany compared to a RMSE of 0.13 Wh/Wp for singlePV systems. Besides the forecast accuracy, also the specificationof the forecast uncertainty is an important issue for an effectiveapplication. An approach to derive weather specific confidence intervalsis presented that describe the maximum expecteduncertainty of theforecast.