Study On Slope Stability Analysis And Displacement Prediction Model Of Ion Type Rare Earth Mine

The ion type rare earth mines in southern Jiangxi adopt the in-situ leaching method,which can effectively protect the ecological environment of the mining area without excavating the topsoil and damaging the vegetation of the mining area.However,after investigation,it is found that the geological damage of rare earth mines in southern Jiangxi Province is still relatively serious in recent years.The main reason is that the in-situ leaching process has changed the mechanical properties of rare earth mine soil,destroyed the soil structure,affected the stability of mine slope,and even caused landslide,collapse and other geological disasters to a certain extent,which has a huge impact on the local people’s life,so it has a great impact on the rare earth mine The stability analysis of soil mine slope is very important.In this paper,the zudong rare earth mine in Longnan City is taken as the research object,and the engineering geological survey data of the mining area is obtained through field investigation.Undisturbed soil is taken for indoor test,and the physical and mechanical parameters of the mining soil are determined,which provides data basis for the subsequent three-dimensional modeling of the slope.MIDAS / GTS is used to analyze the slope under natural conditions and rainstorm conditions,and different engineering parameters are obtained The safety factor of the slope under the two conditions is calculated,and the differences of displacement,stress and strain of the slope under the two conditions are compared.This paper introduces the on-line monitoring system of mine stope landslide,and uses BP algorithm and GA-BP algorithm to predict and analyze the displacement data of monitoring points in the study area,so as to judge the accuracy of the prediction model.(1)The moisture content,density,initial void ratio,permeability coefficient,elastic modulus,compression modulus,cohesion and internal friction angle of undisturbed soil were measured by laboratory test.The test results show that the moisture content of topsoil is larger than that of other soil layers,the density of soil layer tends to increase with the depth of soil layer,and the cohesion and internal friction angle of soil layer increase The friction angle is small,the soil layer is sandy but contains certain clay particles,and the permeability coefficient of Longnan rare earth mine is large.(2)Through MIDAS / GTS software to analyze the slope stability under two different conditions,it is concluded that the safety factor of the slope under natural conditions is larger,and the instability phenomenon will not occur;under rainstorm conditions,the safety factor is smaller,the internal stress and effective stress of the slope decrease,the shear stress and pore stress increase,the equivalent strain and plastic strain increase significantly,and the plastic zone is obvious,so the slope is prone to collapse There is no landslide phenomenon.(3)The on-line monitoring system of mine stope landslide is constructed.The surface displacement of the slope is monitored by the crack meter,the deep deformation of the mountain is monitored by the displacement meter,the stress of the supporting engineering is monitored by the stress meter,and the seepage is monitored by the osmometer.When a certain monitoring data exceeds the specified critical value,the system will give an early warning.(4)In order to avoid the disadvantages of BP-ANN such as slow convergence speed and low learning efficiency,GA is used to optimize the weights and thresholds of BP neural network,and then the slope displacement is predicted respectively.The prediction results show that the performance of GA-BP neural network algorithm is higher than that of BP algorithm,and the prediction accuracy of GA-BP neural network algorithm is higher.GA-BP neural network algorithm can be used to optimize the monitoring system and predict the slope displacement.