运营期盾构隧道结构病害的自动化检测与三维可视化

Automated 3D Inspection and Visualization of Structural Defects in Shield Tunnel Linings

在地铁盾构隧道运营过程中,隧道结构病害的高效检测是保障隧道安全稳定的重要手段,其中隧道结构变形和衬砌表观病害是常见的两大类结构病害检测对象.为了弥补传统人工检测方法低效率性和强主观性的缺陷,本文提出了一种基于移动激光扫描技术的盾构隧道结构病害自动化检测和三维可视化方法.该方法利用隧道的激光点云数据首先进行去噪和生成灰度图像的预处理.去噪后的点云数据通过椭圆拟合的方式分别计算出隧道横断面的收敛变形值、椭圆度以及环间错台值;同时利用生成的灰度图像通过深度学习的算法对衬砌表面的渗漏水和剥落病害进行自动化识别和量化.为了将自动检测生成的隧道结构病害结果实现三维可视化,本研究提出了基于激光点云数据的曲面重构方法,通过三维游戏引擎Unity 3D软件的二次开发,实现了隧道内部环境的第一人称三维漫游展示.本文提出的方法成功应用于某地铁盾构隧道的运营检测中,验证了方法的可行性,为地铁盾构隧道的结构病害数字化管理提供了全新的视角和技术手段.

Carrying out efficient inspection on structural defects is vital for ensuring the safety and stability of metro shield tunnel operation.Tunnel deformation and lining defects are two common types of tunnel inspection targets.This study proposes an automated 3 D inspection and visualization method for shield tunnel structural defects using mobile laser scanning technology, which mitigates the shortcomings of low efficiency and high subjectivity in the traditional manual inspection approach.The proposed method firstly preprocesses the tunnel point cloud data using denoising techniques and subsequently generates grayscale images of the tunnel linings.The denoised point cloud data is then used to compute the convergence, ellipticity value, and longitudinal dislocation of the tunnel transverse section through the ellipse fitting method.The generated grayscale images are used to detect and quantify the water leakage and spalling defects on the lining surface using a deep learning model.To realize the 3 D visualization of the structural defects, a 3 D surface reconstruction algorithm is proposed to generate a 3 D tunnel model.The first-person 3 D roaming of the tunnel linings is realized by implementing the 3 D model into the Unity 3 D platform.The proposed method was successfully applied to the inspection of a metro shield tunnel and produced sound performance.This study provides a new perspective and technical means for the digital management of structural defects in metro shield tunnels.