Application of support vector machine method in FY4A ground solar radiation product correction
Lin Xuefei1, Yang Lin2, Tian Pengju3, Zhu Jun4, Li Guangyi3
1. Zunyi Meteorological Bureau of Guizhou Province, Zunyi 563000; 2. The Information Center of Guizhou Province, Guiyang 520001; 3. Guizhou Ecological Meteorology and Agrometeorology Center, Guiyang 550002; 4. Guizhou Climate Center, Guiyang 550002
Abstract：In order to obtain high-precision and high spatiotemporal resolution solar radiation spatial distribution data,a correction model for FY-4A inversion of total radiation values based on support vector machine(SVM) method was established using daily solar radiation data from seven ground radiation stations in Guizhou from March 1,2018 to April 30,2020.The results indicate that there is a significant difference between the pre correction FY-4A inversion total radiation and ground solar radiation observations,and the FY-4A inversion radiation values are generally higher,indicating a significant correlation between the two.After using SVM regression correction model to correct the ground solar radiation of FY-4A,the correlation coefficient between the two increased by 0.047,the average absolute error decreased by 50.00%,the average error improved by 87.49%,and the root mean square error decreased by 63.03%.The degree of dispersion after correction has decreased,and the correlation has significantly enhanced.The research results aim to better apply FY-4A ground solar radiation data to industries such as agricultural zoning,photovoltaic power generation,and building design.
2025, Vol. 42 
