New Combined Models for Estimating Daily Global Solar Radiation Based on Sunshine Duration in Humid Regions: A Case Study in South China

Junliang Fan,Xiukang Wang,Lifeng Wu,Fucang Zhang,Hua Bai,Xianghui Lu,Youzhen Xiang

DOI: https://doi.org/10.1016/j.enconman.2017.11.085

IF: 10.4

2018-01-01

Energy Conversion and Management

Abstract:In the present study, two single (T1 and T2) and eight combined (C1-C8) sunshine-based empirical models for estimating daily horizontal global solar radiation were compared, and four new combined sunshine-based models (P1-P4) were developed. The accuracy and applicability of these models were further evaluated in South China with humid subtropical and tropical climates influenced by the Asian monsoon as a case study. For this purpose, daily meteorological data during 1966-2000 from 20 radiation stations were used for model calibration and the data from 2001 to 2015 were used to validate the models. These models were evaluated and compared based on coefficient of determination (R-2), root mean square error (RMSE), normalized root mean square error (NRMSE), mean bias error (MBE) and a global performance index (GPI). The single sunshine-based model T-2 was recommended for R, estimation with acceptable accuracy when using only sunshine duration data. In terms of combined sunshine-based models, the newly proposed model P3 was the optimum model for daily global solar radiation estimation when sunshine duration, air temperature and precipitation were available. The proposed complex model P4 was recommended for estimating daily global solar radiation with an excellent accuracy when vapor pressure deficit and relatively humidity were further available. Generally, the proposed combined models based on sunshine duration can be applied for daily global solar radiation estimation with high accuracy in humid regions of China and maybe elsewhere with similar climate.