Damage detection of metro tunnel structure through transmissibility function and cross correlation analysis using local excitation and measurement

• Published in: Mechanical systems and signal processing Vol. 60-61; pp. 59 - 74
• Main Authors: Feng, Lei, Yi, Xiaohua, Zhu, Dapeng, Xie, Xiongyao, Wang, Yang
• Format: Journal Article
• Language: English
• Published: Elsevier Ltd 01.08.2015
• Subjects: Construction; Cross correlation; Damage detection; Excitation; Mathematical analysis; Mathematical models; Metro tunnel; Structural damage; Structural health monitoring; Subway tunnels; Subways; Transmissibility function; Structural health monitoring; Damage detection; Transmissibility function; Metro tunnel; Cross correlation
• ISSN: 0888-3270; 1096-1216
• DOI: 10.1016/j.ymssp.2015.02.007

In a modern metropolis, metro rail systems have become a dominant mode for mass transportation. The structural health of a metro tunnel is closely related to public safety. Many vibration-based techniques for detecting and locating structural damage have been developed in the past several decades. However, most damage detection techniques and validation tests are focused on bridge and building structures; very few studies have been reported on tunnel structures. Among these techniques, transmissibility function and cross correlation analysis are two well-known diagnostic approaches. The former operates in frequency domain and the latter in time domain. Both approaches can be applied to detect and locate damage through acceleration data obtained from sensor arrays. Furthermore, the two approaches can directly utilize structural response data without requiring excitation measurement, which offers advantages in field testing on a large structure. In this research, a numerical finite element model of a metro tunnel is built and different types of structural defects are introduced at multiple locations of the tunnel. Transmissibility function and cross correlation analysis are applied to perform structural damage detection and localization, based on simulated structural vibration data. Numerical results demonstrate that the introduced defects can be successfully identified and located. The sensitivity and feasibility of the two approaches have been verified when sufficient distribution of measurement locations is available. Damage detection results of the two different approaches are compared and discussed. •This paper proposes an approach to perform structural damage detection and localization in metro tunnel engineering, which compares the transmissibility function and cross correlation analysis.•The feasibility and sensitivity of the proposed approach have been validated for different tunnel structural damage scenarios.•The proposed experimental strategies represent a localized scheme for excitation and measurement, where only a few vibration measurements very close to the impact excitation are needed.