隧道开挖支护质量3DZI检测技术及应用研究

为解决钻爆法隧道施工过程中隧道围岩及开挖支护质量难以快速、完整、准确检测的难题,通过对三维重建与全景展开图技术的融合与改进,提出基于3DZI技术(3D Reality and Deep Zoom Image)的隧道开挖与支护质量检测方法。通过用参数化流线与迹线方程构建网格化的设计轮廓曲面,建立三维轮廓与二维展开平面之间的映射关系,实现三维实景与平面全景展开图的信息协同与辨识;通过网格曲面法线方程与三维实景模型求交,实现对空间距离测量和体积估算,并生成净空偏差云图和断面图;通过求交修正后的网格曲面实现对图像无视差矫正拼接,生成高质量的全景展开图,用于还原记录围岩和初支表面情况,实现对表面物体长度和面积的测量。结果表明:该方法设备成本低且作业灵活,仅需在现场布置若干控制点后自由设站拍摄照片,可实现隧道围岩地质编录、超欠挖与初支工程量等量化检测,以及隧道塌方体积与大变形量的快速测定,并可揭示各项作业效果之间的因果关联性。工程现场应用表明:该方法检测结果能直观全面地反映隧道开挖支护效果及围岩地质情况,同时获得超欠挖分布规律和初支材料用量信息,可为隧道施工开挖与支护质量控制、动态设计提供可靠的信息化检测结果和依据。

Quality Detection for Tunnel Excavation and Support Based on 3DZI

It is difficult to quickly, completely and accurately detect the quality of tunnel excavation and support during the process of drilling and blasting tunnel construction. To solve this problem, this paper proposes a method for detecting the tunnel excavation and support quality based on three-dimensional (3D) reality and Deep Zoom Image (3DZI). The method is performed through the improvement and integration of 3D reconstruction and panoramic image stitching technology. Firstly, a design profile surface grid is built using the parameterized streamline and trace equation, and the mapping relationship between the 3D contour and the 2D layout plane can be established. Further, the space distance measurement and volume estimation can be realized through the intersection of the normal equation of the grid surface and the 3D real scene model, and the clearance deviation can be generated in the form of a nephogram and a sectional diagram. After intersection correction, the corrected surface grid can be used for image rectification and panorama mosaics without parallax. Finally, a high-quality panoramic image can be generated and used to comprehensively record the information on the tunnel excavating face and the initial support surface, and to measure the length and area of the surface objects. The proposed method has the advantages of low equipment cost and flexible in operation. This method only needs the free-standing photos, which were acquired by setting up several control points on site, to perform the geological identification of tunnel surrounding rock, quantitative detection of over- and under- excavation, and quality detection of initial support, rapid determination of tunnel collapse volume, and large deformation measurement of initial support. Meanwhile, the causal relationship between various operations can be revealed. The field application of this method shows that the acquired detection results can directly and comprehensively reflect the supporting effect of tunnel excavation and the geological conditions of surrounding rock; additionally, the distribution of over- and under-excavation and the used volume amount of initial support materials can be obtained. These results indicate that the proposed method can provide reliable detection results for the quality control and dynamic design of tunnel excavation and support construction.