increases in AE entropy and covariance of the strain field. The monitoring data of each physical field provide early warning asynchronously: the early-warning time of AE entropy is the earliest while that of the AE quiet period is the latest. Membership functions are constructed taking early-warning results of monitoring data in various physical cases as bodies of evidence, and their critical values are determined using normal distribution. Multi-sensor data are fused using the Dempster-Shafer (D-S) evidence theory to realize time-varying prediction and provision of early warning with graded probabilities for tunnel failure based on collaborative multi-sensor data fusion. Article Highlights• The development of cracks and their initiation mechanisms of the tunnel model are explored. • The monitoring data of each physical field provide early warning of tunnel asynchronously. • Multi-sensor data are fused to realize time-varying prediction and warning with graded probabilities of tunnel failure. Keywords Tunnel • Crack propagation • Acoustic emission • Failure warning • Multi-sensor data fusion List of symbols Ω Covariance matrix S dTotal number of strain components Abstract Biaxial compression tests were conducted on circular tunnels that constructed in sandstone to reveal the evolution of macro-cracks, the strain field, and acoustic emission (AE) parameters during deformation of tunnels. An early-warning method for tunnel failure based on multi-sensor data fusion was studied. The results indicate that tensile cracks first initiate at the arch haunch of the tunnel; then shear cracks occur on both sides of the opening, forming two asymmetric V-shaped grooves; finally, tensileshear cracks on both sides of the tunnel propagate to the vault, triggering brittle failure of the tunnel. The strain concentration zones appear earlier than cracks in the tunnel, and the location and range of strain concentration zones cover the crack-propagation paths. Before the failure of tunnels, a series of precursor events occur, including an AE quiet period, a significant, rapid drop in the AE-based b value, and abrupt
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.