Font Size: a A A

Study On Deformation Prediction Of Taojiapo Landslide Based On Information Gain Method

Posted on:2021-02-24Degree:MasterType:Thesis
Country:ChinaCandidate:W ZhangFull Text:PDF
GTID:2480306467464404Subject:Architecture and Civil Engineering
Abstract/Summary:PDF Full Text Request
Due to the influence of water level fluctuation and water storage in the Three Gorges Reservoir area,the frequency of geological disasters in the reservoir area is increasing.The Yangtze River is the golden waterway in China,and the Three Gorges reservoir is located in an important section,so ensuring the production and living safety of the people in the reservoir area is the most important.Monitoring and early warning is one of the effective and important means to deal with landslide geological hazards.In this paper,according to the landform characteristics of taojiapo landslide,the stability of taojiapo landslide is analyzed in sections,the spatial development law of landslide inducing factors is mined by ArcGIS software,the influence degree of inducing factors on landslide is calculated by information gain method,and the most unfavorable inducing factors are screened.In order to ensure the credibility and accuracy of the calculation results of information gain method,chi square test is adopted to test the relevance of the data.Then the geo studio modeling software is used to simulate the change of rainfall and reservoir water level.Finally,the deformation prediction of taojiapo landslide is carried out by analyzing the change of stability coefficient.The main research steps and conclusions are as follows:(1)According to the geological and monitoring materials collected by taojiapo,the mining research is carried out to extract useful information,and the information is screened and analyzed to determine the segmentation standard,and then the taojiapo landslide is divided into sections.(2)On the basis of segment,the stability of each segment is calculated.And through the comparison of unstable and stable sections,stable and stable sections,unstable and unstable sections.Different factors that may induce landslide deformation are obtained,which are called induced factors in this paper.(3)The inducing factors are analyzed by ArcGIS,and the spatial structure of landslide is described.In the study area,the elevation changes obviously and fluctuates greatly.In general,it is high in the north and low in the south,high in the East and low in the west,and there is a sudden change of elevation in some parts;the gradient changes greatly,with the highest gradient of 71 °,and the gentle slope of 20 ° to 35 ° in most areas;the overall gradient fluctuates obviously,mainly following the irregular distribution of the mountain fluctuation.(4)Using the information gain algorithm,the information gain of each inducing factor is calculated respectively,and the results are compared and analyzed.It is concluded that the change of rainfall and reservoir water level are the two factors that have the greatest influence on the landslide deformation.(5)The most unfavorable factors are modeled by geo Studio software.The maximum daily rainfall in 20 years rainfall is taken as the simulated rainfall,and the variation of landslide stability coefficient is simulated by taking the different duration as the variable,and the rule is summarized.In the aspect of reservoir water level simulation,taking one year as a cycle,the simulation of reservoir water level is a rising and falling process.In the process of simulation,the change of stability coefficient is found,and finally the law is summarized.(6)Based on the analysis of all aspects,the prediction analysis is carried out,the prediction scheme is determined,and the early warning conclusion is given.The research results have reference significance for the prediction of similar landslides.In the research process of this paper.The idea of data mining is adopted,and information gain method is used to select the main influencing factors from multiple data,and the main inducing factors are determined by simulation calculation.Based on the study of the traditional single landslide overall early warning and regional meteorological early warning,the deformation prediction scheme for taojiapo landslide is analyzed and summarized..
Keywords/Search Tags:taojiapo landslide, information gain method, ArcGIS, geo-studio modeling, Deformation prediction
PDF Full Text Request
Related items