Probability prediction of solar irradiance in the tropic using ensemble forecasting

Abstract

As photovoltaic (PV) power generation systems become more widespread instability of the electric power grids with PV connection is becoming an issue. For appropriate management of the grids, probability prediction of solar irradiance is proposed. Lagged Average Forecast (LAF) method is used for ensemble forecasting. The 72-hour ahead forecasting of solar irradiance is operated in Thailand once a day, and it contains intra-day, next-day and 2-day ahead forecastings. The ensemble forecasting has three ensemble members. The accuracy of intra-day forecasting is higher than the other members, and it is employed as the most probable value of the forecasting. The relation between spread and forecasting error is analyzed. From the result, the confidence intervals of the predictions are derived for arbitrary reliability. The probability prediction is performed with the most probable value and the confidence intervals. The interval changes its width due to spread changes and captures the observation in it.