Research On Instability Early Warning Criterion And Deformation Prediction Of Soft Rock Cutting Slope In Mountainous Area

In recent years,as the climate changes,various extreme weathers have followed.In particular,frequent and continuous rainstorms have caused various geological disasters,especially slope instability and damage,which has seriously threatened people’s lives and property.Taking mountain roads as the research object,this paper has carried out research on instability warning and deformation prediction of soft rock cut slopes in mountainous areas,which has important theoretical and practical significance.On the basis of on-site investigation and survey and collection and analysis of relevant data,through indoor geotechnical tests,theoretical analysis and numerical simulation and other research methods,the soft rock cutting slope deformation instability threshold warning criterion and deformation prediction model are studied,and combined The engineering example verifies the reliability and effectiveness of the early warning prediction method.The research results can provide a certain reference for the design and construction of related projects.The main research contents are as follows:(1)Laboratory tests were carried out on the rock and soil of the soft rock slope of the engineering,and basic physical and mechanical parameters such as elastic modulus,Poisson’s ratio,cohesion and internal friction angle were obtained.(2)According to the meteorological and hydrological data of the project location,numerical simulation method is used to analyze the slope stability change law considering the effect of rainfall infiltration under different rainfall conditions;combined with the theory of seepage and stress coupling,the displacement changes and The relationship between the safety factors;combining the analysis of the slope deformation stage and the classification criteria of the warning level to establish the slope deformation displacement threshold warning criterion.(3)Based on the analysis and summary of the applicability and limitations of the gray system theory and BP neural network in slope deformation prediction applications,the GM(1,1)and BP neural network parallel integrated model and series integrated model are proposed,and Compile their related calculation programs based on MATLAB software.And through the comparative analysis of typical historical slope examples,it is found that the accuracy of the parallel integrated model and the series integrated model in slope deformation prediction is greater than that of the single prediction model,and the prediction accuracy of the series integrated model is better than that of the parallel integrated model.(4)Combining engineering examples,comparing and analyzing the actual displacement data on site and the deformation prediction results using the comprehensive prediction model,the results show that the model is effective and reliable.At the same time,the current stability of the slope is determined based on the warning threshold combined with the surface displacement monitoring data The performance status is judged,the slope stability change trend is analyzed through the comprehensive model prediction results,and an early warning system based on prediction is established.