Prediction of solar radiation intensity in clear sky based on AOD estimation model

Abstract

Atmospheric aerosol is one of the most important factors that cause the random variation of solar radiation intensity. In view of the problem that the atmospheric aerosol optical depth (AOD) is difficult to obtain real-timely with high accuracy, BP neural network method is adopted to estimate the AOD based on PM2.5 concentration, PM10 concentration, air temperature and air relative humidity that from air quality monitoring station; then, take AOD and precipitable water as main change parameters, REST2 model is simplified to calculate direct solar radiation intensity and scattered radiation intensity in clear sky. Experimental results show that the proposed method for predicting solar radiation intensity in clear sky is of high precision and easy to be realized.