Rail transit shield tunnel deformation detection method based on cloth simulation filtering with point cloud cylindrical projection

Abstract

Subway shield tunnels are important components of urban infrastructure. To maintain the safety of a tunnel during operation, tunnel deformation must be detected in a timely manner. The large quantities of metal brackets, electrical equipment, and other ancillary facilities installed along the lining of subway shield tunnels, along with bolt holes in the shield ring plate, significantly affect the detection accuracy of tunnel deformation. To increase deformation detection accuracy, a filtering method for tunnel ancillary facilities based on cloth simulation filtering with point cloud cylindrical projection is proposed. In this method, a cylindrical projection of the original point cloud of the tunnel cross-section is created. In this projection, each tunnel lining point corresponds to a projection onto the horizontal plane and the tunnel ancillaries are expanded onto the lining surface. The cloth simulation filtering algorithm is then used to separate the point clouds of the tunnel lining and ancillaries. The filtered tunnel section curves are fitted with an ellipse model based on the direct least squares method and the deformation of the tunnel is analyzed using the ellipse parameters. The repeatability values of tunnel measurements of the overall and local deformation are 0.41 and 0.66 mm, respectively. Indicating that the proposed filtering method can effectively obtain a tunnel lining point cloud from the original point cloud of the tunnel and meet the accuracy requirements of subway tunnel deformation detection.