基于机器学习的FY4A反演北方冬季降水的研究  

Research on FY4A Inversion of Northern Winter Precipitation Based on Machine Learning

作　　者：刘辉 LIU Hui(Meteorological Data Center of Inner Mongolia Autonomous Region,Hohhot,Inner Mongolia 010000,China)

机构地区：[1]内蒙古自治区气象数据中心,内蒙古呼和浩特010000

出　　处：《自动化应用》2023年第21期15-18,共4页Automation Application

摘　　要：现有的静止卫星降水产品在内蒙古冬季表现较差,为提高卫星反演冬季降水的准确性,本文介绍了基于机器学习的静止卫星FY4A反演北方冬季降水,使用分类算法、回归算法分别构建了降水区域模型和降水量模型,并将静止卫星遥感数据输入降水区域和降水量联合模型得到最终降水反演结果。评估结果显示,与FY4A降水产品相比,模型反演降水产品的各项指标均有所提高,效果较好。The existing geostationary satellite precipitation products perform poorly in winter in Inner Mongolia.In order to improve the accuracy of satellite inversion of winter precipitation,this paper introduces a machine learning based geostationary satellite FY4A to invert winter precipitation in northern China,using the classification algorithms and precipitation model to constructed the precipitation region model and precipitation model,and input the geostationary satellite remote sensing data into the precipitation region and precipitation joint model to obtain the final precipitation inversion result.The evaluation results show that the indicators of the model invert precipitation has improved compared with the FY4A precipitation product,with good results.

关 键 词：XGBoost 机器学习 卫星反演 FY4A 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]