A forecast system assimilating weather radar data
By extracting physical constraints from radar data and assimilating multi-source data, the problems of response delay and misjudgment of non-meteorological noise in traditional numerical forecasting systems are solved, achieving highly reliable meteorological decision support and improving the scientificity and accuracy of business decisions.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- YUNHAI ZHICHUANG (JIANGSU) TECHNOLOGY CO LTD
- Filing Date
- 2026-04-23
- Publication Date
- 2026-07-07
AI Technical Summary
Traditional numerical weather prediction systems suffer from response delays and misjudgments of non-meteorological noise in industry applications, making it difficult to meet the timeliness and operability requirements in business decision-making. In particular, in scenarios such as traffic control, logistics scheduling, or energy trading, existing machine learning-based forecasting systems lack the ability to identify the physical consistency of radar observations, resulting in poor generalization ability and affecting the credibility of decisions.
By constructing a radar data acquisition module, a physical knowledge construction module, a physical constraint extraction module, and a multi-source data assimilation module, hierarchical physical identification and suppression of ground clutter and super-refractive echoes are achieved. Combined with multi-source data, dynamic coordination fusion is performed to generate a highly reliable industry risk index.
It has improved the credibility and data reliability of meteorological decision-making, enhanced the quality control of meteorological information for supervisory purposes, solved the problem of model output violating atmospheric physical laws, improved the scientificity and accuracy of decision-making in high-reliability commercial scenarios, and ensured the continuity and comparability of meteorological risk assessment.
Smart Images

Figure CN122085419B_ABST
Abstract
Citation Information
Patent Citations
Lake thermodynamic process simulation method and simulation system
CN117875193A
Solar radiation space-time prediction method and system based on physical information constraint and neural network
CN120951237A