Hydrological model real-time correction method and system based on ensemble learning algorithm

• Main Authors: WANG XINYI, ZHANG YE, SHANG SHOUWEI, XIE SONG, WANG LEIZHI, LI XITING, LI LINGJIE, LIU JINXIA, NIU KAIJIE, CUI TINGTING, LIU YONG, SU XIN
• Format: Patent
• Language: Chinese; English
• Published: 18.08.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; MEASURING; METEOROLOGY; PHYSICS; TESTING

The invention discloses a hydrological model real-time correction method and system based on an ensemble learning algorithm, and the method comprises the following steps: collecting the information of a research region, obtaining a basin type and hydrological characteristics, and constructing a coupling flood forecasting model; aiming at the coupled flood forecasting model, screening and classifying correction methods to obtain at least two types of correction methods, and correcting the hydrological model by adopting each correction method to obtain a predicted flow sequence; a random forest ensemble learning model is constructed and trained; constructing an evaluation index set, evaluating the forecasting effect of the random forest ensemble learning model, and if a preset evaluation value is reached, outputting a hydrological model based on an ensemble learning algorithm; and performing flood prediction by adopting the hydrological model based on the ensemble learning algorithm to obtain a hydrological seq