Paper Conference

Proceedings of Building Simulation 2021: 17th Conference of IBPSA

     

Effectiveness of cloud cover on solar radiation prediction using artificial neural network algorithm

Hany Gaballa, Soolyeon Cho
North Carolina State University, United States of America

DOI: https://doi.org/10.26868/25222708.2021.30304
Abstract: Solar radiation data is highly desirable in various areas, such as agriculture, PV industries, and building performance analysis. There are no commercial tools available for real-time solar radiation prediction using only readily available data such as temperature and humidity. The purpose of this paper is to test if solar radiation can be predicted within acceptable uncertainties using only temperature and humidity, based on ASHRAE Guideline 14-2014. It is also discussed how much “cloud cover” affects solar radiation prediction when using ANN algorithms. Three climate locations are tested in Arizona, Washington, and North Carolina. Studies show that the prediction of solar radiation using ANNs depends primarily on the correlation between cloud cover and global solar radiation, rather than on the cloud cover ratio.
Keywords: Cloud cover, Solar radiation prediction, Artificial Neural Network, Real-time building performance, Real-time weather file
Pages: 1397 - 1403
Paper:
bs2021_30304