Forecast of ensemble power production by grid-connected PV systems

The contribution of power production by PV systems to the electricity\nsupply is constantly increasing. Anefficient use of the fluctuating\nsolar power production will highly benefit from forecast information\non the expected powerproduction. This forecast information is necessary\nfor the management of the electricity grids and for solar energy\ntrading. This paper will present and evaluate an approach to forecast\nregional PV power production. The forecast quality was investigated\nfor single systems and for ensembles of distributed PV systems. Due\nto spatial averaging effects the forecast for an ensemble of distributed\nsystems shows higher quality than the forecast for single systems.\nForecast errors are reduced to an RMSE of 0.05 Wh/Wp for an ensemble\nof the size of Germany compared to a RMSE of 0.13 Wh/Wp for single\nPV systems. Besides the forecast accuracy, also the specification\nof the forecast uncertainty is an important issue for an effective\napplication. An approach to derive weather specific confidence intervals\nis presented that describe the maximum expecteduncertainty of the\nforecast.