Groundwater Detection Of Reservoir Landslide And Its Influence On Landslide Stability

Reservoir landslide seriously threatens the economic and social development of the reservoir area,and has a negative impact on the safety of human life and property.Groundwater is one of the main factors inducing the deformation and failure of reservoir landslide.However,the existing soil moisture monitoring methods have the problems of long time-consuming and large investment,and it is difficult to effectively monitor groundwater for a long time.In view of this problem,Baijiabao landslide is taken as the research object,and the high-density resistivity method is applied to investigate the influence of reservoir water level fluctuation.The systematic research is carried out by means of field investigation,indoor physical model test,numerical simulation,field monitoring and big data analysis.The relationship between the deformation and failure of reservoir landslide and the distribution of groundwater is revealed,and the prediction model of water content of multiphase composite materials based on high-density resistivity method is proposed to form the integrated technology of groundwater monitoring and numerical analysis of reservoir bank landslide.The research results can provide reliable theoretical and technical support for effective monitoring of landslide deformation,landslide control and protection mode innovation in the Three Gorges Reservoir area.The main conclusions are as follows :(1)Taking Baijiabao landslide in the Three Gorges Reservoir area as the research object,laboratory tests are carried out to study the electrical conductivity characteristics of landslide soil under different influencing factors.Through correlation analysis,it is shown that compared with dry density and void ratio,the changes of moisture content and temperature in landslide body have greater influence on soil resistivity.(2)Based on the topography and geomorphology characteristics of Baijiabao landslide,the indoor landslide physical model test is carried out,and a hybrid model of long-short-term memory neural network model and Gaussian process regression model based on Bayesian optimization algorithm is proposed,namely Bayesian LSTM-GPR.The model is evaluated and verified by multiple evaluation indexes.The results show that the proposed method can obtain higher precision water content interval prediction value and more reliable probability prediction value in a short time.(3)The water content prediction model of multiphase composite materials based on high density resistivity method was applied to the field test of Baijiabao landslide to obtain the distribution law of groundwater in the landslide.The results show that the groundwater level of Baijiabao landslide near the reservoir is obviously changed.In the part far from the reservoir,the water content of landslide soil changes slowly.The prediction results of the proposed method are in good agreement with the field sampling values,indicating that the method can be applied to the actual monitoring of landslide in reservoir area.(4)The Baijiabao landslide model is established by using the commercial finite element software.Adoption of the DP criterion equivalent to Mohr Coulomb,the strength parameters c and φ、μ and E are reduced together,and the influence of groundwater distribution on landslide stability is explored by integrating a variety of instability criteria.The results show that groundwater has a great influence on the safety performance of landslide in reservoir area,and the monitoring of groundwater cannot be ignored in the process of landslide monitoring and early warning.