Accurate nowcasting of cloud cover at solar photovoltaic plants using geostationary satellite images.

ABSTRACT: Accurate nowcasting for cloud fraction is still intractable challenge for stable solar photovoltaic electricity generation. By combining continuous radiance images measured by geostationary satellite and an advanced recurrent neural network, we develop a nowcasting algorithm for predicting cloud fraction at the leading time of 0-4 h at photovoltaic plants. Based on this algorithm, a cyclically updated prediction system is also established and tested at five photovoltaic plants and several stations with cloud fraction observations in China. The results demonstrate that the cloud fraction nowcasting is efficient, high quality and adaptable. Particularly, it shows an excellent forecast performance within the first 2-hour leading time, with an average correlation coefficient close to 0.8 betwe

SUBMITTER: Xia P 

PROVIDER: S-EPMC10787801 | biostudies-literature | 2024 Jan

REPOSITORIES: biostudies-literature