Research on Machine Learning-Based Precipitation Forecast Correction Technology for Shaoguan City
TANG Penghui
HUANG Shaozhong
ZHANG Weibiao
ZHAO Juliang
LI Guanyi
Abstract：To accurately forecast heavy precipitation during the Dragon Boat Water period and provide precise precipitation data for ecological environment monitoring,a forecast correction technique suitable for heavy precipitation in Shaoguan City during this period is proposed.Based on clarifying the terrain and con-fluence characteristics of the study area,the temporal and spatial features of heavy precipitation during the Dragon Boat Water period are extracted.Using terrain,confluence characteristics,and the spatiotemporal features of heavy precipitation as reference,a target function for heavy precipitation forecast correction is es-tablished.The weighted integration method is applied to fuse the solutions of the target function,obtaining the final corrected values for heavy precipitation forecasts during the Dragon Boat Water period.Experiments demonstrate that the proposed method achieves high correction accuracy and can provide accurate heavy pre-cipitation forecasts for Shaoguan City's ecological environment monitoring.This study provides an effective technical approach for improving the accuracy of heavy precipitation forecasts,which is of great significance for disaster prevention and mitigation,as well as ecological protection in the region.
Keywords：weather forecastheavy precipitationmachine learningforecast correctionsingular val-ue decompositionShaoguan City
Publication Date：2025-06-20
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：5( 6-10 )