Research On Deformation Detection Of Metro Tunnel In Operation Period Based On Laser Scanning Technology

By the end of 2020,a total of 7,978.19 km of urban rail transit lines have been opened in 45 cities in mainland China,of which subway lines are 7108.49 km long,accounting for 89.09%.With the increase of subway lines,periodic deformation detection of subway tunnels becomes very important in order to ensure the safe and reliable operation of subway.Compared with traditional detection methods,laser scanning technology has obvious advantages of rich measurement data,high measurement accuracy and high measurement efficiency,and has gradually become a new technology for subway tunnel detection.In view of the actual requirements of deformation detection of subway tunnels in operation period,this paper has carried out research on tunnel point cloud data processing and deformation analysis method based on laser scanning technology.In the process of point cloud data preprocessing,gearing toward the problems of tunnel point cloud gigantic quantity of data,high redundancy in some areas and the density distributing evenly,this paper adopts the way of space voxel extract adjacent points,downsizes the point cloud efficiently on the premise of meet the demand of deformation analysis.Compared with the point cloud data simplification algorithm basing on octree,the density distribution of point cloud is more uniform after simplification.And the noise point in the point cloud data is classified by means of multi-level algorithm,first the outliers are removed by the method of spatial density statistics,and then the cloth simulation filter algorithm is introduced to separate the ground points,through the experimental analysis shows that cloth grid is set to 0.5 m,height classification threshold of 0.3 m can achieve a good separation effect.and finally adopting iterative least square algorithm ellipse fitting and random sampling consistency ellipse fitting to remove the mixing points,both methods have achieved good results.In the research of tunnel deformation analysis method based on point cloud data,aiming at the problem of lack of auxiliary data in tunnel point cloud data processing during operation period,this paper proposes a spatial cylinder iterative fitting algorithm to extract tunnel axis,to the point cloud data of the tunnel line,the easement curve and circular curve of the single linear or comprehensive extract linear axis,through the simulation experiment accuracy verification,the central axis anchor point extraction error of the complex line is about 3mm,which can meet the precision requirements of engineering applications.After the extraction of the central axis and section,Tunnel ellipticity detection,converging deformation detection,cross-section wrong station detection,track detection and boundary detection have been carried out for deformation analysis.The results of ellipticity detection were in good agreement with those processed by commercial software Amberg Tunnel 2.0.Finally,in order to improve the intelligence and visualization of data processing,a tunnel point cloud data processing experimental platform has been built based on the open source point cloud data processing framework in this paper.The platform has four functional modules of mass point cloud data management,visualization,tunnel deformation detection and results report export,and realizes the whole process of tunnel deformation detection data processing based on laser scanning technology.