Slope Stability Analysis And Prevention Measures Based On Strength Reduction Method And Neural Network

With the mining of open pit mines,slope problems will inevitably occur,which will bring potential safety hazards to the staff and equipment of the mining area.Therefore,this paper took Xinjiang Jilangde open pit coal mine as an example,combined with open pit engineering geological data and on-site exploration results,and found out The engineering geological conditions in the study area,and then the limit equilibrium method was used to conduct anti-sliding stability analysis on 9 sections of the excavation site.The stability coefficient is analyzed,and then the BP neural network is used to predict the stability coefficient of the homogeneous slope and evaluate the prediction results by training 112 sets of sample data.Finally,three treatment schemes are proposed,and a further three-dimensional model is further established.Compare the three governance options and make recommendations.The main research results of this paper are as follows:(1)Divide the Jilangde coal mine excavation into 9 sections,and use the limit equilibrium method to evaluate the 9 sections.According to the evaluation results,it can be concluded that,under natural conditions,the stability factor Fs of the 1-1' section of the west of the excavation of the Jilangde open-pit mine does not meet the 1.3 required by the code.(2)The Midas GTS NX software was used to simulate the 1-1 'section of the west side of the excavation from three angles: maximum shear strain cloud map,horizontal displacement cloud map,and slope stability coefficient.Through simulation,the potential sliding surface of the slope and the displacement under rainfall were further determined,and the calculation results of the limit equilibrium method were also confirmed.(3)A total of 112 sets of slope sample parameters were collected from the literature,and the slope stability coefficient was predicted by using the method of BP neural network,and compared with the actual value.From the comparison results: Bayesian regularization is adopted The prediction accuracy of the BP neural network method is better than the LM method and the quantized conjugate gradient method.(4)Three treatment plans are proposed for the treatment of the west side of the excavation site.The engineering support measures include single-row anti-slide piles,prestressed anchor cables and anchor rods,and shotcrete surface layer on the mudstonesurface.1-1 'section to build a pseudo three-dimensional model,and compare the three schemes from three angles of maximum shear strain,horizontal displacement,slope stability coefficient,etc.The comparison results show that the slope management can be achieved only by using anti-slide piles Purpose,but for the long-term limited stability of the slope,this article recommends the joint use of anti-slide piles and anchor cables to reinforce the west side of the mining site.