Study On Slope Monitoring And Early Warning Method Based On Ground SAR High Frequency Region Data

Mine slope instability is one of the main factors restricting the safe production of open-pit mines in China.In recent years,with the continuous deepening of open-pit mines in China,the slopes of many open-pit mines have been continuously increased and steepened,forming high and steep slopes,and the risk of landslides has become higher and higher,which has caused serious threats to the staff in the mining area.Therefore,it is very important to study the time,location and sliding range of mine landslide in China to ensure the safety of mine production and the life and health of workers.In this study,Liba Gold Mine in Gansu Province was taken as the research area.Ground-based SAR was used to monitor the slope of the mine stope all day and all weather,and the radar deformation data was obtained in real time.At the same time,combined with terrain registration technology,the radar deformation data is matched with UAV aerial survey data to realize the visualization of 3D deformation cloud image.In addition,in order to accurately locate the landslide location,delineate the landslide range and predict the landslide time,this thesis studies the traditional early warning model and method based on the deformation data after radar registration,and analyzes the influence of slope height,angle,geological structure and cumulative precipitation on the early warning accuracy of slope landslide.The main conclusions showed that:(1)A three-parameter landslide early warning model analysis method based on deformation velocity,cumulative deformation displacement and critical sliding area is proposed.Combined with the factors such as slope height,angle,cohesion,friction angle and rainfall,the short-term and long-term landslide early warning system is established,and the blue,yellow,orange and red four levels of early warning are determined.(2)FLAC3D numerical simulation software is used to solve the displacement,velocity and other parameters of the slope to verify the applicability and accuracy of the three-parameter early warning analysis model.MATLAB software is used to optimize the area parameters,and a polygon area estimation method is proposed,which effectively improves the accuracy of sliding area estimation(3)The characteristic deformation laws of single point and regional deformation data in different warning ranges are different.It is found that the reciprocal velocity curves in different periods have obvious characteristic convergence points in different regions,and the convergence time is different.The characteristic convergence points of different regions can be used to warn the slope landslide disaster in advance.At the same time,the speed reciprocal warning method is improved,and the applicability and accuracy of the improved speed reciprocal warning method are verified,highlight the readability,readability and regularity of the monitoring data,and the prediction results are accurate.