Shaking table tests on seismic response and damage mode of tunnel linings in diverse tunnel-void interaction states

• Published in: Tunnelling and underground space technology Vol. 77; pp. 295 - 304
• Main Authors: Xin, C.L., Wang, Z.Z., Gao, B.
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.07.2018; Elsevier BV
• Subjects: Acceleration; Anisotropy; Arches; Compression tests; Damage patterns; Earthquake damage; Fracture mechanics; Materials science; Seismic response; Seismology; Shake table tests; Shaking table test; Tunnel engineering; Tunnel linings; Tunnels; Void-lining interaction; Voids; Shaking table test; Void-lining interaction; Tunnel engineering; Seismic response
• ISSN: 0886-7798; 1878-4364
• DOI: 10.1016/j.tust.2018.03.010

•A shaking table-based method was put up to study void-lining interaction in seismic dynamic states.•The tunnel models behavior with / without voids behind the linings in the shaking table tests were compared. The voids between the primary and the secondary linings for tunnel structures can be regarded as the worst contact state, which is deemed as serious tunnel quality defect. Recently, researchers have attempted to study tunnel linings behavior with the voids mostly in static state but these sort of defective tunnel linings are easier to destroy by earthquake. A shaking table-based method to determine the behavior of tunnels subject to the voids behind linings can be the ideal method to study their interaction in seismic dynamic states. This paper presents results from a series shaking table tests on scaled tunnel models with and without the void on the lining crown under increasing seismic intensities excitations. Details of experimental setup and procedures are described first and then the test results are presented. The test results are discussed based on the acceleration amplification factors, tension-compression strains, and damage patterns. The comparison shows that the peak acceleration on the outer surface of lining crown with a void is greater than without a void after inputting the peak ground acceleration (PGA) beyond 0.4 g regardless of the excitation directions. Each monitoring point on lining models without a void is mainly in the reasonable stress state and the most adverse strain state generates on the arch springing. There is nearly no strains difference between lining models in tight and loose contact states while the input PGA grows from 0.2 g to 0.4 g. However, plenty of tension strains appear and increase from the crown to the sidewall of lining model with a void with the input PGA increasing from 0.6 g to 1.0 g. Accordingly, multiple longitudinal cracks along axis of lining model are generated from the crown to the sidewall but three annular cracks on outer surface of lining crown should be focused attention. Finally, it is found that tunnel models behavior with and without voids behind the linings in the shaking table tests compared favorably with that in seismic damage instances.