Regional landslide risk early warning method based on rainfall threshold

• Main Authors: LIU DONGSHENG, PENG SHUANGQING, ZHANG TIANYOU, CHEN GANG, WANG LIPING, JING XILIANG, LUO JUN, NING CHAOLIE, ZHANG WEI, LUO WENWEN
• Format: Patent
• Language: Chinese; English
• Published: 04.08.2023
• Subjects: ALARM SYSTEMS; 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; ORDER TELEGRAPHS; PHYSICS; SIGNALLING; SIGNALLING OR CALLING SYSTEMS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention relates to a regional landslide risk early warning method based on a rainfall threshold value, and belongs to the field of disaster early warning. The method comprises the following steps: S1, establishing an information amount-random forest IV-RF and information amount-BP neural network IV-BPNN landslide susceptibility prediction coupling model based on information amount and negative sample selection; s2, establishing a continuous probability rainfall threshold model based on multi-function fitting; and S3, carrying out regional landslide risk early warning based on the susceptibility and a continuous probability rainfall threshold value. According to the method, the IV is coupled with the RF model and the BPNN model to obtain two coupling models, and optimization is performed in the two models to obtain a higher-precision landslide susceptibility value of the research area. And the landslide susceptibility value and the continuous probability value are coupled to obtain a continuous probabili