青藏高原低涡月度生成频数预测模型构建方法、青藏高原低涡预测方法及装置
By constructing a monthly frequency prediction model for low-pressure systems on the Tibetan Plateau by region and time period, and by using a two-way stepwise regression algorithm and meteorological factor feature screening, the problem of monthly-scale prediction of low-pressure systems on the Tibetan Plateau in existing technologies has been solved, and reliable predictions for 1-3 months have been achieved, improving prediction accuracy and reliability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- CHINESE ACAD OF METEOROLOGICAL SCI
- Filing Date
- 2026-06-18
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies cannot make monthly-scale climate predictions for the Tibetan Plateau low-pressure system. The reliable forecast lead time of numerical weather prediction models is less than 15 days, which leads to rapid distortion of forecast results in key elements such as location, intensity, and formation time.
A monthly frequency prediction model for low-pressure vortex formation on the Tibetan Plateau was constructed using a regional and time-segmented approach. Candidate features were screened and extended features were constructed through a bidirectional stepwise regression algorithm. Historical monthly average values of meteorological factors were obtained, and the monthly frequency prediction model was determined based on the model structure, combined with adaptive calibration of monthly coefficients.
It has achieved reliable prediction of the frequency of low-pressure vortex formation on the Tibetan Plateau over a period of 1-3 months, significantly improving prediction accuracy and reliability, capturing spatial heterogeneity and seasonal non-stationarity, and providing a high-precision monthly forecasting tool.
Smart Images
Figure CN122412929A_ABST