Abstract
This disclosure describes techniques for solar power generation forecasting. Multiple camera nodes are deployed along with a threshold number of rooftop solar panels within a geographic region. With appropriate permissions, the camera nodes capture and transmit images of the sky to a server along with location and time information. Each cloud is identified individually, and the height of each cloud is estimated based on its movement. The information for individual clouds is merged to create a sky map of separate layers of clouds based on their height. A consolidated sky map that includes clouds from all heights is transmitted to different solar power generation systems. The sky map with cloud information is utilized to predict future positions of the clouds relative to the local solar power generation system and corresponding cloud cover is determined. Based on the cloud cover, a forecast of solar power generation capacity is obtained.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Sato, Koichi and Sanders, Nicholas, "Real-time Solar Power Generation Forecasting Based on Cloud Cover Estimation", Technical Disclosure Commons, (May 09, 2023)
https://www.tdcommons.org/dpubs_series/5879