Visibility graph analysis on time series of shield tunneling parameters based on complex network theory. (July 2019)

• Record Type: Journal Article
• Authors: Zhou, Cheng; Ding, Lieyun; Zhou, Ying; Skibniewski, Miroslaw J.
• Is Part Of: Tunnelling and underground space technology. Volume 89(2019)
• Page Start: 10
• Page End: 24
• Publication Date: 2019-07
• Subjects: Shield tunneling -- Parameter -- Complex network -- Visibility graph -- Time series -- Metro construction; Tunneling -- Periodicals; Underground construction -- Periodicals; Tunnels -- Periodicals; Underground areas -- Periodicals; 624.193
• DOI: 10.1016/j.tust.2019.03.019
• Languages: English
• ISSNs: 0886-7798

Abstract

Highlights: A visibility graph model implemented in a complex network theory of shield tunneling is proposed. Degree distributions of the reconstructed networks from time series of shield parameters follow the power law. The characteristics of shield tunneling parameters' network are scale-free and small-world. Findings reveal turning points in the time series of shield parameters with vital node analysis. It is tested and validated in the first Yangtze river crossing metro tunnel project in China. Abstract: Shield tunneling operation affects project scheduling, cost estimation, and safety risk during metro construction. Given the importance of this process, tunnel engineers should understand the complexities of shield tunneling parameters. This study addresses the characteristics of shield tunneling parameters by introducing a visibility graph model implemented in a complex network of shield tunneling in metro construction. Time series data are collected for the analysis of six parameters generated in each segment ring tunneling cycle. These parameters are total thrust, torque, penetration rate, rotation rate, advance speed, and working chamber pressure. Bridging time series and the complex network with visibility graph algorithm indicates that all degree distributions of the construct networks follow the power law, establishing the scale-free time series of the parameters. Small-world features and hierarchical structures also indicate that the fluctuation and variance of the parameters are attracted, limited, or affected by a previous shield operation of segment rings. In addition, findings reveal turning points in the parameters' time series with vital node identification based on complex network analysis. Implications relevant for shield operators and managers are proposed to improve shield tunneling performance, efficiency, and safety.