The Research On The Detection Method Of Surface Foreign Body And Section Deformation In Subway Tunnel And The Development Of Experimental Prototype

With the rapid urbanization in China,the subway can well solve the problem of urban traffic jam.However,there are various safety risks in the subway project.So it is urgent to develop a fast and efficient equipment to detecting tunnel’s diseases.Among them,surface cracks and deformation of tunnel section can display the problems of tunnel safety synthetically.At present,the performance of equipment for detecting subway safety is uneven: some equipment is still using relatively traditional manual methods,which can only be used for single point detection of the subway tunnels,but cannot be used for a comprehensive inspection of subway tunnels,and the detection efficiency is also relatively low;some is the imported,or self-developed automatic testing equipment,however,it’s cost is relatively high,and the detection of disease is not complete.This thesis is based on the project of developing a safety photoelectric comprehensive detection system for subway tunnel,the detection system integrates the detection of tunnel wall surface crack and the detection of tunnel section deformation,which can realize the detection of subway tunnel with high efficiency and low cost.At present,the detection system can detect subway tunnels at the speed of 32km/h,and the detection accuracy of tunnel deformation can reach 2mm.The minimum size of the object that can be segmented is 10 mm.This thesis mainly describes the development of hardware and software of the detection system and the processing of 3D point cloud data,the specific research contents of this thesis are as follows:(1)This thesis introduces the causes of diseases of subway tunnels,and then studies and compares the relevant cases at home and abroad.And then,according to the theoretical calculation and practical application scenarios,the safety detection system is designed and the devices are selected,as well as the system data acquisition software is designed.(2)According to the characteristics of this system,a pre-processing method(denoising and compressing)for 3D point cloud data collected by 2D laser scanner is designed and realized.A method based on density clustering is proposed and implemented to segment 3D point cloud data of tunnel objects,the results of segmentation are compared with that of region growing segmentation,and the advantages of the proposed method in this system are proved.(3)In this thesis,the main points of tunnel deformation analysis based on 3D point cloud are discussed,and a method of synthetically judging the deformation of tunnel based on both macroscopic(the ovality of the tunnel)and microscopic(the shape variable of tunnel cross section).is proposed and implemented.(4)Finally,the developed experimental prototype was used to collect data from the simulated tunnel built by the partner,proving that the developed detection system met the detection requirements.And the applicability and the effectiveness of the proposed algorithm is verified by the collected data.