Assessment of the diffuse-fraction estimation in China using the binary BRL model

Diffuse and direct radiation distribution are crucial for the design and construction of photovoltaic power systems. However, due to the high cost and relatively complex maintenance, there are relatively few observation stations for diffuse and direct radiation in China, while there are numerous stations for total radiation, making it easier to obtain total radiation data. In this study, based on the Boland Ridley Laurent (BRL) model, the atmospheric radiation transfer theory and Taylor series expansion were utilized to establish a BRL model relating the diffuse fraction to the clear-sky index and atmospheric optical thickness. Training datasets comprising radiation data at three different time scales, namely minute, 10-min, and hourly, were used to obtain the mathematical model parameters for the diffuse fraction with respect to the clear-sky index and atmospheric optical thickness through nonlinear fitting, resulting in binary BRL model. Validation datasets were created using solar radiation data from observation stations in regions with distinct climatic characteristics. The BRL model was employed to calculate the diffuse fraction, statistical analyses were conducted on the correlation coefficient, mean bias deviation (MBD), root mean squared deviation (RMSD), and t-statistic between measured and computed diffuse fraction values. The results show that the diffuse fraction values computed by the BRL model exhibited correlation coefficients above 0.8, MBD within ±0.2, RMSD within 0.25, and minimum t-statistic as low as 0.1074. This study provides a new direction for constructing diffuse radiation model, and further research should incorporate more extensive radiation data to investigate the model’s applicability across different regions.