A reservoir flood forecasting method fusing empirical model and AI algorithm
By integrating empirical models and AI algorithms into reservoir flood forecasting, utilizing watershed topography and soil type to divide response units, and identifying statistical mutations in real time and updating the error mapping network and observation noise covariance matrix online, the forecasting error problem caused by abrupt changes in the upstream discharge rules of cascade reservoir groups was solved, achieving stability and accuracy in flood forecasting.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- JIANGXI ACAD OF WATER RESOURCES (JIANGXI PROVINCE DAM SAFETY MANAGEMENT CENT JIANGXI PROVINCE WATER RESOURCES MANAGEMENT CENT)
- Filing Date
- 2026-06-15
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies, under conditions of abrupt changes in the discharge rules upstream of a cascade reservoir group, are unable to suppress non-stationary jumps in empirical model errors and achieve rapid online adaptive correction of error prediction models without relying on a large number of historical abrupt change samples and long-term complete data.
By integrating empirical models with AI algorithms, response units are divided using watershed topography and soil type, establishing empirical relationships between water storage and discharge, and combining online change point detection algorithms and error mapping networks to identify statistical abrupt change moments in real time, and update the error mapping network and observation noise covariance matrix online, thereby achieving rapid correction of flood forecasts.
This improves the stability of flood forecasting for cascade reservoir groups under non-stationary scheduling scenarios, reduces reliance on long-sequence complete data, and ensures the continuity and accuracy of flood forecasting.
Smart Images
Figure CN122414734A_ABST