基于大语言与数据-机理的作物产量预测方法及系统
By combining large language models and data-driven models, and utilizing allometric growth constraints and asynchronous assimilation techniques, the accuracy and robustness issues of existing crop yield prediction methods under spatial adaptability and extreme climate conditions are solved, achieving high-precision and robust crop yield prediction.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- NORTHWEST A & F UNIV
- Filing Date
- 2026-05-14
- Publication Date
- 2026-07-17
AI Technical Summary
Existing crop yield forecasting methods have limitations in terms of spatial adaptability, generalization ability, data assimilation technology, and system interaction threshold, resulting in insufficient forecast accuracy and robustness, especially under extreme climatic conditions.
We employ a large language model for semantic analysis of multi-source data, combined with parallel inference driven by both mechanistic and data-driven models. Through asynchronous assimilation strategies and allometric growth constraints, we dynamically adjust the model fusion weights to achieve high-precision and robust prediction of crop growth status.
It improves the accuracy and robustness of crop yield forecasting, solves the problems of root-shoot ratio imbalance and forecasting bias under extreme weather conditions in traditional methods, reduces the complexity of system operation, and achieves high-precision and robust yield forecasting.
Smart Images
Figure CN122198266B_ABST