| Rail transportation plays an increasingly important role in the urbanization process and economic development,and the construction of subway tunnels has also received great attention,accompanied by the urgent need for monitoring the relative deformation of tunnels.The traditional monitoring of subway tunnels is carried out by placing a certain number of control points in the tunnel,setting up a total station to observe the coordinates of the control points,and then manually detecting the disease.Not only is it inefficient,but it does not reflect the whole picture.3D laser scanning technology has the outstanding advantages of large data volume,high accuracy and fast speed,and is increasingly important in tunnel settlement and relative deformation detection.Mobile 3D laser scanning technologies are more suited to the narrow,slender environment of tunnels than stationary ones,and their principles and working methods differ considerably.The current three-dimension laser scanning technique focuses on the convergence analysis of the sections,and the relative deformation is determined by comparing the shapes of the sections at different times.No study of the relative deformation of tunnels and tracks along mileage has been seen.In this paper,a method for detecting and locating the linear relative deformation of subway tunnels is proposed with reference to the detection method of track irregularity,which identifies and extracts the feature points of interest from the cross-sectional point cloud,and combines wavelet decomposition and wavelet energy spectrum methods to determine the wavelength,position and amplitude information of the relative deformation at a specific frequency.This paper presents the current research and application of tunnel deformation and 3D laser scanning technology.The use of 3D laser scanning technology in the construction of subway tunnels,the principle of acquiring point clouds,the characteristics of the scanning instrument and the composition of the scanning trolley are presented.The advantages of mobile scanning technology in tunnel scanning are elucidated,as well as its differences in calculating point cloud coordinates and target corrections compared to stationary scanning.Cross-sectional interception of point clouds,from which a series of feature points of interest are identified and extracted to form feature curves to present the distribution of mileage along specific parts of the tunnel.In order to detect the relative deformation,the local relative deformation fluctuation index of the linear structure is proposed,the wavelet base function and the number of decomposition layers are searched for the appropriate wavelet decomposition of the feature curve,and then a fast Fourier transform is performed on the obtained detail and approximate signals,and the spectra of all signals are analyzed to determine the wavelength of the relative deformation.The smoothed pseudo Wigner-Ville distribution(or SPWVD)can be seen as the distribution of the signal energy in the time frequency domain,from which the SPWVD of the detail signal with the maximum time resolution for the feature frequency is calculated,the mileage distribution curve corresponding to the feature frequency is extracted,and the detection threshold is set according to the Laida criterion to determine the position of the relative deformation.Find the corresponding waveform from the corresponding position of the approximate signal and determine the magnitude of the relative deformation.The validity and accuracy of the algorithm is verified using simulation data,and finally,the algorithm is processed using real point clouds of a subway tunnel in Wuhan City.The results show that the algorithm is useful for the detection of relative deformation of tunnels and can be used as a new method for deformation detection for reference. |
PDF Full Text Request