Research On Tunnel Deformation Monitoring Method Based On Point Cloud Slice Extraction Technology

With the progress of science and technology and the rapid development of urban rail transit,large-scale construction of metro projects has been carried out in major and medium-sized cities throughout the country in recent years.At the same time,many high-rise buildings have been built near Metro tunnels.During the construction process,both metro tunnels have been affected and deformation phenomena such as settlement or displacement have taken place to varying degrees,which will result in cracks accompanied by water seepage.Situation occurs.If these deformations are not paid attention to,they may accumulate more and less in the long run,which will destroy the safety and stability of existing tunnels.Serious safety accidents will result in heavy losses to society and economy.Based on the requirement of deformation monitoring,three-dimensional laser scanning technology,with its unique advantages of speed,accuracy and non-contact data acquisition,has become another new technology leading representative in the field of deformation monitoring.However,with the continuous updating of this technology,the amount of point cloud data obtained is huge,and the post-processing is time-consuming and laborious.This technology is facing enormous challenges.Therefore,this paper focuses on the introduction of a point cloud slicing algorithm based on total least squares(TLS)and weighted principal component analysis(WPCA),which is used to segment the original point cloud of the tunnel section,so as to extract relevant information for deformation analysis.At the same time,taking an existing subway tunnel in Wuhan as an example,this paper discusses the application of point cloud slicing technology in tunnel deformation monitoring.The main work of this paper is as follows:(1)Based on the total least squares method and using the pre-processed tunnel data,the bottom plane and the top plane of the tunnel are fitted successively and expressed in the form of formulas.(2)Using weighted principal component analysis method,the normal vectors of point clouds on the inner wall of the channel are obtained and the data sets are established.On this basis,the datum level is obtained.Considering the principle that the angle between the normal vector of datum level and the normal vector of point cloud in tunnel inner wall should not exceed 2 degrees,the best datum level suitable for point cloud slicing is selected.The calculation results show that the cosine value of the angle between the normal vector of the best datum and the normal vector of the corresponding points meets the requirements.(3)The point cloud data collected at two hours interval are substituted for calculation,and the center of gravity coordinates of the tunnel slices are extracted,the distance between the bottom and the top is calculated,so as to carry out the deformation monitoring of the existing tunnel.(4)Based on the idea of curve fitting in numerical analysis,a space circle curve fitting experiment is used to fit the axis of tunnel bend,and the axis of tunnel slice is extracted successfully.The overall deformation trend of tunnel is judged by comparing the deformation of tunnel axis with the two-stage data.On the basis of the above four points,this paper provides an idea for monitoring the slice deformation of tunnels,and prospects the future development of this research.