陕西水利Issue(6)：9-11,16,4.
基于Self-Organizing Maps回归算法的黄河流域降水量空间预测研究
Spatial Prediction of Precipitation in the Yellow River Basin Based on the Self-Organizing Maps Regression Algorithm
摘要
Abstract
A spatial prediction model for precipitation in the Yellow River Basin was developed based on the Self-Organizing Maps(SOM)regression algorithm.Using precipitation observation data from 305 meteorological stations in 2020,combined with geographical and environmental factors such as elevation,slope,aspect,and NDVI,the SOM model parameters were optimized through a grid search method.The results indicate that the SOM model successfully captured the spatial heterogeneity of precipitation in the Yellow River Basin,achieving high prediction accuracy(R2=0.83,RMSE=47.6 mm).Precipitation exhibited a decreasing trend from the southeast to the northwest,ranging from 135 mm to 1171 mm.High-value areas(>900 mm)were mainly distributed in the southeastern region,mid-value areas(500 mm～800 mm)were located in the central region,and low-value areas(<400 mm)were concentrated in the northwestern region.This study provides an effective new approach for spatial precipitation prediction.关键词
Self-Organizing Maps/降水量/黄河流域/空间预测Key words
Self-Organizing maps/precipitation/Yellow River Basin/spatial prediction分类
大气科学引用本文复制引用
刘文婷,白明照,李凤云..基于Self-Organizing Maps回归算法的黄河流域降水量空间预测研究[J].陕西水利,2025,(6):9-11,16,4.基金项目
南岸灌区水土资源优化配置及智慧管控关键技术研究与示范项目(NSK202201) （NSK202201）
