Seismic vulnerability of circular tunnels in sand

Abstract

The paper is focused on the assessment of seismic fragility curves for circular tunnels under moderate to severe earthquakes with the aim of improving the reliability of the risk assessment of underground infrastructural networks. Non-linear two-dimensional dynamic analyses were performed on different tunnel and soil configurations by using the finite-difference method implemented in Flac2D software. The applied input motions were selected considering their amplitude and frequency content variability. The response accelerations and predominant frequencies computed at ground level, above the tunnel, were compared with the corresponding free-field results to distinguish the effects attributable to the tunnel presence from those due to the site amplification. Tunnel safety was assessed through fragility curves, taking into account the dependence of tunnel lining bending resistance on the axial force variation during the earthquake. The more effective intensity measure was investigated correlating the tunnel performance to peak ground accelerations and peak ground velocities computed at the ground level and at the bedrock depth, respectively. The resulting fragility curves showed a satisfying matching with the empirical ones, generated on the basis of the observed seismic damage on tunnels.

Developments in Seismic Vulnerability Assessment of Tunnels and Underground Structures

“…Furthermore, de Silva et al [72] proposed a series of analytical fragility curves for the assessment of circular tunnels in sand, using the results of cloud analysis, involving 2D nonlinear dynamic analyses of selected soil-tunnel configurations, carried out in FD code FLAC 2D. In particular, a 10 m diameter circular tunnel, with lining thickness t = 0.50 m, was examined.…”

“…In the simplest case, a linear elastic model [53,54], or approaches that employ the equivalent linear approximation in simulating the effect of the nonlinear response of ground under seismic shaking, were employed [63][64][65][66][67][68][69][70][76][77][78][79]. Nonlinear models that employ a Mohr-Coulomb yield criterion were also used widely [44,52,55,56,60,72,73,82]. In some studies, these models were calibrated by adjusting the stiffness and damping of the examined ground, based on results of separate 1D soil response analyses, aiming for a more realistic simulation of the ground response under lower earthquake-induced strains [52,65,71,73].…”

“…Regarding lining behavior, most studies referring to mountain tunnels in rock or bored tunnels in alluvial employed a linear elastic model [51][52][53][54][55][56]60,61,65,[70][71][72][73]79]. The damage index and the damage state thresholds were defined using a separate section analysis for the examined liners, accounting for the effect of axial loading on the capacity of the liners in a simplified manner.…”

“…Moreover, the linear elastic model employed by these studies does not replicate cracking and/or yielding phenomena on the liner, associated with the response at higher levels of ground shaking. As stated in Section 3.1.2, de Silva et al [72] proposed an analytical method to account for the effect of axial load time dependency on the resistance bending moment of the lining and, hence, on the computation of the damage to capacity ratio. However, the definition was made for one damage state.…”

Combination of lining strengthening and buffer layers for soft and hard rock tunnels junction subjected to seismic waves

“…The serious damages such as the lining dislocation and collapses of rocks were observed in Baiyunding Tunnel of G123 national highway (Shen et al 2014;Tsinidis et al 2020). An extensive work has been done on the seismic response and seismic effect of tunnels under different seismic conditions (Fabozzi et al 2018;Silva et al 2020). However, there are few studies considering the seismic behaviours of tunnels occurring in the junction of soft and hard surrounding rocks.…”