Scene flood forecasting method and system based on Xinanjiang and deep learning coupling model

• Main Authors: HE JIANBING, ZHANG YE, LI CHEN, WANG YINTANG, WANG LEIZHI, YUN ZHAODE, ZHU RONGJIN, SHAN YUSHU, LI LINGJIE, WANG QIN, NIU KAIJIE, CAI WENTING
• Format: Patent
• Language: Chinese; English
• Published: 30.08.2022
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention relates to a session flood forecasting method and system based on a Xinanjiang and deep learning coupling model, and the method comprises the steps: collecting and preprocessing the geographic information data and rainwater condition data of a pre-research region, and constructing a Xinanjiang three-water-source hour-scale hydrological model; obtaining flood initial parameters of the Xinanjiang three-water-source hour-scale hydrological model based on the geographic information data and the rainwater condition data, calibrating and checking the Xinanjiang three-water-source hour-scale hydrological model, and screening out a preset number of session floods through the Xinanjiang three-water-source hour-scale hydrological model, obtaining an output variable and an intermediate state variable of the flood of each session; a deep learning coupling model matched with the Xinanjiang three-water-source hour-scale hydrological model is constructed, a Xinanjiang and deep learning coupling model is formed