一种基于可解释机器学习的蒸发互补关系参数确定方法
By using an interpretable machine learning approach, key driving factors of evaporation complementarity parameters are identified and partitioned, solving the problems of low accuracy and poor global applicability in the determination of evaporation complementarity parameters in existing technologies, and achieving more accurate global evaporation estimation.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHANGJIANG RIVER SCI RES INST CHANGJIANG WATER RESOURCES COMMISSION
- Filing Date
- 2026-03-27
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies, when determining parameters for evaporation complementarity, suffer from low accuracy and poor applicability on a global scale because they neglect spatial variation patterns and the combined influence of multiple driving factors.
By employing an interpretable machine learning approach, we collect global watershed data, construct an enhanced regression tree model, identify key driving factors, partition the data based on feature importance, and establish a model for determining evaporation complementarity parameters, thereby improving global applicability.
It improves the simulation accuracy of evaporation complementarity parameters, especially performing well in extreme climate regions, overcoming the limitations of traditional models in regional adaptability, and providing reliable support for global-scale evaporation estimation.
Smart Images
Figure CN121936306B_ABST