For solar forecasting, the operational numerical weather prediction (NWP) models have been shown to be consistently erroneous at predicting irradiance and thus limited in applicability. Generally, NWP over-predict irradiance, implying an under-prediction of cloud cover. This error can be attributed to several sources: Domain resolution, model physics parameterizations, and inaccurate initial conditions. To address these issues, a high-resolution, cloud-assimilating NWP for solar irradiance forecasting was developed and implemented at the University of California, San Diego, and GL-Garrad Hassan, America, Inc (WRF-CLDDA). Here, model resolution and physics parameterizations were specifically chosen to foster cloud growth and development. Furthermore, a cloud-assimilation system was implemented to directly and accurately initialize clouds into the model. Simulations were performed for 5/1/2011 to 6/30/2011 and irradiance forecasts established for UCSD's dense pyranometer network. This study validates the model's capability of predicting irradiance variability and presents a method of deriving sub-hourly variability statistics from hourly output irradiance fields. Overall, WRF-CLDDA is shown to accurately predict irradiance variability for ramp rates with temporal scales of as fine as 20 min.
展开▼
