Intelligent on-duty early warning method and system fused with meteorological service data
By combining historical prior database, mutual information method, deep learning and Kalman filtering, the problem of difficulty in integrating multi-source data in traditional meteorological warning systems has been solved, efficient and accurate warning of complex meteorological events has been achieved, and the intelligence level of the meteorological warning system has been improved.
Patent Information
- Application Number
- CN202510638713.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-19
- Publication Date
- 2025-09-23
- Estimated Expiration
- 2045-05-19
AI Technical Summary
Traditional meteorological warning systems rely on manual judgment and traditional mathematical models, and it is difficult to effectively integrate multi-source heterogeneous meteorological data, resulting in low forecast accuracy. In particular, it is difficult to achieve efficient and accurate warnings in complex meteorological events.
A method combining historical prior database, mutual information method, deep learning and Kalman filtering is adopted. The mutual information method is used to screen the characteristics of multi-source data, deep learning is used to process severe convection sudden warnings, and Kalman filtering is used to process slow risk warnings, thus realizing dynamic association and efficient fusion of data.
It improves the accuracy and real-time performance of the meteorological warning system, can effectively handle complex and changeable meteorological phenomena, and enhances the intelligence and practicality of meteorological warnings.