| China’s highway tunnel construction in recent years the scale and speed of rapid development,and the number of tunnels in the past 10 years have an annual growth rate of about 10%.Tunnel second lining is the last line of defense for tunnel safety,in which the construction quality of reinforcement directly determines the protective performance of the second lining,standardizing the construction quality of the second lining reinforcement can not only improve the safety performance of the tunnel but also scientifically grasp the safety status of the tunnel and improve the later operation and maintenance capacity of the tunnel."JTGT 3660-2020 Technical Specification for Highway Tunnel Construction" and "JTG F801-2017 Highway Engineering Quality Inspection and Assessment Standard Book I Civil Engineering"both clearly put forward the lining rebar spacing error requirements,the common rebar Common means of detection include geological radar method,electromagnetic induction and other methods commonly used after the completion of construction,the construction process detection basically relies on manual sampling,with limitations,the quality control of the reinforcement construction process is insufficient.Therefore,a method with a wide range,high accuracy,and strong adaptability is urgently needed to achieve the purpose of obtaining rebar data information and extracting rebar spacing with high quality.This paper proposes to apply 3D laser scanning technology to the quality inspection of second-lining rebar construction in order to achieve efficient and accurate quality inspection of the second-lining rebar construction process.In this paper,through systematic indoor model tests,we analyze the multidimensional point cloud characteristies of the rebar network during the second lining construction phase of highway tunnels and propose a directional extraction method for the second lining rebar during the construction phase of highway tunnels,as well as a nearest neighbor mean distance extraction algorithm for different diameter rebar networks.The work carried out in this paper and the main conclusions are as follows:(1)Introduce 3D laser point cloud technology to carry out experimental research on the construction quality inspection model of highway tunnel second lining rebar,including rebar 3D point cloud model acquisition,data processing,and quality parameter analysis to obtain multi-dimensional point cloud characteristics of rebar network.A multi-functional model testing device for tunnel second lining rebar was designed and piloted.A total of 24 test conditions covering parameters such as rebar diameter and spacing were designed,and the spatial morphological and quantitative characteristics of the point cloud of tunnel second lining rebar were analyzed in terms of overall and local characteristics,as well as the spatial morphological and sampling density variations of various rebar tying conditions to analyze their effects on rebar morphology,density,and continuity.The two layers of the rebar network are independent and the continuity of the first layer is better;the first layer of rebar data is semicylindrical and linear,and the second layer of rebar data is V-shaped column surface and linear;the neighborhood density of longitudinal bars,distribution bars,and a facial point cloud is different,with the largest facial point cloud,the second longitudinal bars,and the smallest distribution bars.(2)The point cloud feature identification and algorithm experiments of the second lining rebar network during the tunnel construction period are studied.The Euclidean clustering,DBSCAN density clustering algorithm,and the spectral analysis clustering extraction algorithm proposed in this paper for rebar morphological features are analyzed,and algorithm experiments are carried out to obtain the best parameters of the three algorithms for rebar identification and the advantages and disadvantages of using them for tunnel second lining rebar network identification.Based on the difficulties of tunnel rebar network identification,the above algorithms are optimized to establish a new road tunnel construction period rebar extraction method of "global point cloud traversal-local feature extraction" for directional tunnel second lining rebar network,and finally,the feasibility and superiority of the method are verified.The optimized algorithm can achieve 100%extraction of the first layer of rebar,and the extraction quality of the second layer of rebar is 186%,186%,and 22.4%higher than the three clustering recognition algorithms,and the overall extraction quality of rebar is 31%,31%and 21%higher than the three algorithms,respectively.(3)The theoretical method of rebar spacing extraction is studied and experimentally verified by the spacing extraction algorithm.Based on the data characteristics of the rebar point cloud,the research is carried out from two aspects of the rebar point cloud data as a whole and the fitted feature lines,respectively,and a variety of rebar spacing identification and extraction algorithms are analyzed,and the relevant algorithms are improved by combining the point cloud data characteristics,and a nearest neighbor mean algorithm is proposed for different diameter combination rebar networks,and finally a variety of spacing intelligent extraction algorithms are compared by the algorithm experiments.The accuracy of the algorithm is 99%for longitudinal bar spacing extraction and 97%for distribution bar spacing extraction,which improves the accuracy of longitudinal bar extraction by about 6 times and meets the practical application requirements.(4)The method of this paper is applied to the quality inspection of rebar in the tunnel construction process,and the actual inspection of lining rebar is carried out at the road tunnel construction site by using 3D LIDAR,which verifies the feasibility of the extraction method of the point cloud model of tunnel lining rebar and the effectiveness of the automatic identification method of rebar network spacing parameters.The accuracy meets the lining rebar spacing error requirement,and the detection efficiency is substantially improved compared with manual detection,which can meet the rapid inspection of lining rebar construction quality in tunnels. |
PDF Full Text Request