Abstract The management, health monitoring and maintenance of operational metro tunnels have been gradually becoming digital and intelligent. However, the lack of digital tunnel models often limits the application and development of intelligent maintenance and management systems for metro shield tunnel management and inspection organizations. This paper proposes an intelligent reconstruction method of the digital model of disordered erected segment ring structure in metro shield tunnels based on deep learning and machine vision, uses high-definition pictures of the inner surface of the tunnel lining obtained by inspection vehicles to intelligently identify and automatically classify the tunnel segment features (bolt holes), and then automatically infers the layout pattern of the tunnel segment rings according to the distribution characteristics of the bolt hole groups, thus achieving rapid reconstruction of the tunnel digital model by combining with the actual tunnel alignment. The application case in a certain metro tunnel shows that the proposed method is applicable to shield tunnels with irregularly and staggered erected segments, and can achieve the intelligent reconstruction of the digital model of the metro shield tunnel with 100% accuracy.
Abstract：
The management, health monitoring and maintenance of operational metro tunnels have been gradually becoming digital and intelligent. However, the lack of digital tunnel models often limits the application and development of intelligent maintenance and management systems for metro shield tunnel management and inspection organizations. This paper proposes an intelligent reconstruction method of the digital model of disordered erected segment ring structure in metro shield tunnels based on deep learning and machine vision, uses high-definition pictures of the inner surface of the tunnel lining obtained by inspection vehicles to intelligently identify and automatically classify the tunnel segment features (bolt holes), and then automatically infers the layout pattern of the tunnel segment rings according to the distribution characteristics of the bolt hole groups, thus achieving rapid reconstruction of the tunnel digital model by combining with the actual tunnel alignment. The application case in a certain metro tunnel shows that the proposed method is applicable to shield tunnels with irregularly and staggered erected segments, and can achieve the intelligent reconstruction of the digital model of the metro shield tunnel with 100% accuracy.
ZHANG Chun ZHOU Yuxuan LI Dengpeng
.Intelligent Reconstruction of the Digital Model of Metro Shield Tunnels with Disordered Erected Segment Ring Structure[J] MODERN TUNNELLING TECHNOLOGY, 2022,V59(1): 80-86
2022, Vol. 59 