Study On Monitoring And Early Warning Of Slope Deformation In Mine With Gently Inclined Weak Intercalation

Slope deformation and instability damage has the characteristics of continuity and instantaneousness,which poses a safety threat to production and construction,so slope monitoring and early warning is indispensable.In this paper,the slope of a limestone mine with soft interlayer in Sichuan area is taken as the research object,and the variation characteristics of the slope displacement deformation curve under this special geological condition are relatively different than that of the typical slope,so a variety of prediction models are used to study the slope displacement deformation.Based on the prediction results of the displacement model,the engineering analogy method and the fuzzy analytic hierarchy method are used to comprehensively judge from both qualitative and quantitative aspects,and establish early warning and forecasting indicators combining displacement threshold,displacement rate angle,displacement acceleration and macroscopic deformation characteristics,and carry out early warning and forecasting of slopes with weak sandwich mines.The main research results are as follows:1.Through the analysis and on-site investigation of the relevant data of the Huangshan mining area in Sichuan,the occurrence conditions,geological structure and external influencing factors affecting the stability of the slope of Laoyingzui were comprehensively analyzed.The analysis of material composition,exploration of formation genesis and physical and mechanical properties of the soft interlayer in the mining area showed that the geological conditions showed strong nonlinear characteristics in slope displacement deformation,and displacement prediction was difficult.2.Based on time series,slope displacement is predicted,and gray theory,BP neural network and genetic algorithm are used to model and analyze respectively.Through the GM(1,1)prediction model,the model correction was carried out by the residual method,and the optimal modeling sequence m=20 was obtained,with a maximum relative error of 7.7%.The BP neural network model can show the overall evolution mechanism of slope deformation and failure in the prediction results,and the maximum relative error is 5.2%.The GA-BP neural network prediction model optimizes a single algorithm,weakens the defects of a single algorithm,and reduces the maximum relative error to1.5%,indicating that the GA-BP prediction model has the best effect in predicting the displacement deformation of the slope of mines with weak sandwich layers.3.The influence factor of slope sliding was determined by engineering analogy method and analytic hierarchy method,and the similarity of the studied slope was calculated by weight calculation.Through the collection of data on previous landslides,a similarity ratio database was established to find the two landslides with the highest similarity.Then,the fuzzy analytic hierarchy method was used to correct the membership degree,and the displacement thresholds of the three deformation stages of the Laoyingzui slope were 1.0mm/d,1.7mm/d and 21.0mm/d,respectively.Combined with the macroscopic deformation characteristics,displacement acceleration and displacement rate angle,the average displacement acceleration value is 0.019mm/d~2,and the displacement rate angle A value is within±0.2,and then the early warning and forecasting grading index of the slope with weak sandwich mine is established,and the Laoyingzui slope is judged to be in the initial deformation stage.