Risk division and prediction method of mountain torrent disaster based on gis-neural network integration

Abstract

The invention relates to a mountain torrent disaster risk division and prediction method based on GIS (geographic information system)-neural network integration. The method includes the steps: S1 mining the association relationship between a risk factor and a risk grade in a mountain torrent disaster by the aid of association rules, identifying the risk factor and building a quantitative mountaintorrent disaster risk evaluation index system; S2 determining risk and vulnerability index system by an analytic hierarchy process and the weight of the system to generate feature layers; S3 stackingmountain torrent disaster risk and vulnerability distribution layers by ArcGIS to obtain a mountain torrent disaster risk distribution diagram; S4 performing clustering by an ISO maximum likelihood method, merging regions from bottom to top and performing qualitative analysis from top to bottom to form mountain torrent disaster risk division; S5 analyzing non-linear relationships among evaluationindexes, the risk grade and disaster data by an Elman neural network, and building a mountain torrent disaster risk evaluation and loss prediction model. The problem of spatial scale uncertainty in mountain torrent disaster evaluation in a changing environment can be solved, and the method can be widely used for evaluating mountain torrent disaster risks.

Description

technical field [0001] The invention relates to the field of mountain torrent disaster prevention and control, and more specifically, to a method for zoning and predicting mountain torrent disaster risk based on GIS-neural network integration. Background technique [0002] my country is a mountainous country, and the mountain area accounts for about 2 / 3 of the country's land area. The complex terrain and geological conditions, the climatic characteristics of frequent rainstorms, the dense population distribution and the influence of human activities lead to frequent occurrence of mountain torrent disasters. my country's mountain torrent disasters present the overall characteristics of wide range of influence, high frequency of occurrence, short duration of disaster, large degree of damage and significant regional differences. According to the statistics of the "National Mountain Flood Disaster Prevention and Control Planning Report", the watershed area of ​​my country's hil...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art, provide a kind of mountain torrent disaster risk zoning and prediction method based on GIS-neural network integration, build the mountain torrent disaster risk assessment and loss estimation model, solve the mountain torrent disaster assessment under the changing environment The problem of spatial scale uncertainty in

Images