Suffolk Reporter

A recent study has improved solar energy forecasting by taking into account the types of clouds that impact solar irradiance predictions. Researchers from Brookhaven National Laboratory, along with collaborators from Stony Brook University, Nanjing University of Information Science and Technology, and the National Renewable Energy Laboratory, have advanced solar forecasting science.

Lead researcher Yangang Liu noted the study's reliance on "thanks to the decade-long, high-quality data collected by the U.S. Department of Energy’s Atmospheric Radiation Measurement (ARM) Program." The data, which spans from 2001 to 2014, was critical for assessing solar forecasting models' performance under different cloud conditions.

The research highlights eight cloud types, including cumulus, stratiform clouds, and cirrus, in their influence on solar irradiance predictions. Models informed by physics and cloud-radiation dynamics were evaluated against real-world observations from the ARM South Great Plain (SGP) Central Facility site.

According to the study, models performed best with weak convective clouds such as cirrus, and worst with strong convective clouds like deep convective clouds. Shinjae Yoo emphasized, “By categorizing clouds into stratiform, weak, and strong convective types, we were able to identify where our models performed best and where they needed improvement.” Yoo pointed out particular complexities in forecasting under conditions of deep convective clouds due to their dynamic and unpredictable nature.

Further details can be found on the AI Innovation Institute website through a story by Ankita Nagpal.