Abstract：The article designs and implements a warning support system for short-term heavy precipitation warning business needs at grassroots meteorological stations.The system adopts a hierarchical architecture design and constructs a precipitation feature recognition model based on the fusion of multi-source meteorological data.It uses machine learning algorithms and fuzzy comprehensive evaluation methods to achieve intelligent recognition and early warning of short-term heavy precipitation.The application verification at the Zhuhai Meteorological Observatory in Guangdong Province shows that the system's early warning accuracy reaches nearly 80％,with an average early warning of 22 minutes.The accuracy of early warning for heavy precipitation processes with hourly rainfall exceeding 50 mm can reach over 90％.The research results provide a practical and efficient short-term heavy precipitation warning solution for grassroots meteorological stations,which is of great significance for improving the warning service capability.
邓丽霞, 林珊珊. 面向基层气象站的短时强降水预警支持系统研究[J]. 气象水文海洋仪器, 2026, 43(1): 62-64.
Deng Lixia, Lin Shanshan. Research on shortterm heavy precipitation warning support system for grassroots meteorological stations. Meteorological Hydrological and Marine Instrument, 2026, 43(1): 62-64.
2026, Vol. 43 
