[M-ZZ40] Renewable energy and earth science

[MZZ40-P02] Cloud observation and solar irradiance estimation using all sky imagers in Taiwan

*Shih-Sian Yang¹, Hsin-Chih Lai^2,3, Chen-Jeih Pan¹, Li-Wei Lai³, Min-Chuan Hsiao^2,3, Wan-Ju Liu³ (1.Department of Space Science and Engineering, National Central University, Taiwan, 2.Department of Green Energy and Environmental Resources, Chang Jung Christian University, Taiwan, 3.Environmental Research and Information Center, Chang Jung Christian University, Taiwan)

Keywords:All sky imager, Cloud cover, Solar irradiance, Solar power

The sun emits radiative energy much more than humans consume over the globe. Therefore solar power is considered the most promising among all kinds of renewable energy. However, clouds can easily cover the sun, reducing solar irradiance and generated power suddenly and rapidly. For that reason, cloud observations are critical for nowcasting solar irradiance and further power generation. Recently, the Central Weather Bureau established a network of 17 all sky imagers (ASIs) in Taiwan. This study presents the initial results of cloud cover observation and solar radiance estimation using this ASI network. Color recognition techniques are used to distinguish clouds from the sky, and the cloud cover of the sky is calculated accordingly. Most ASIs within the network were installed beside an automatic weather station. Pyranometer observations are available among these stations, and the pyranometer data enable us to correlate cloud cover with solar irradiance. Besides, the cloud base height can be estimated from all sky images using either dual ASIs or sole ASI methods. The cloud distribution in an all sky image is converted to a two-dimensional cloud map through cloud height information. The cloud map demonstrates the areas covered by clouds. Further, a time series of cloud maps shows the motion and evolution of clouds, providing crucial information for solar irradiance nowcasting. At last, the comparisons between the results of this study and other observations will also be made in this presentation.