Coupling effects of surface building and earthquake loading on in-service shield tunnels

Abstract

In a dense urban environment, an in-service shield-tunnel is very likely to be subjected to additional loading from later-built surface structures, which could couple with the earthquake effects and aggravate the seismic damages of the tunnel structures. The main objective of this study aims at investigating the coupling effects of later-built surface building and earthquake loading on in-service subway shield tunnels. To this end, a series of numerical studies are carried out, employing a Finite Element model in which the segment joints and connecting bolts are simulated. The surface structure is assumed to be a 10-storey three-span reinforced concrete frame on a raft foundation. Some critical indexes, including lining deformation, dynamic lining forces, joint displacement, as well as the mechanical response of the joint bolt are used to evaluate the coupling effects. As a comparison, the responses without the surface building and those with tunnel ling simplified as equivalent continuous rings are also investigated. The results reveal that the coupling effects lead to much higher seismic response of the tunnel structure. The location of the surface building plays an important role on the seismic responses, which increase and then decrease with an increase in the distance s that is measured from the center of the surface building to the tunnel axis.

Introduction

Metro tunnels constitute a critical component in the modern transportation infrastructure to alleviate the traffic pressure in urban areas, including those with active earthquake activities. After the 1995 Kobe earthquake, in which severe damages or collapses were observed for subway stations and tunnels [1], [2], the seismic resistance of underground structures became a critical issue in their designs and rehabilitations. Over the last two decades, many investigators have carried out experimental (e.g. [3], [4], [5], [6], [7], [8]), numerical (e.g. [9], [10], [11], [12], [13]), or analytical studies (e.g.[14], [15], [16], [17]) on the seismic responses of underground structures. In most of these studies, only the underground structures were considered, while the effects of other structures in the tunnel vicinity were neglected.

However, the in-service subway shield tunnels are usually influenced by newly-built surface structures in dense urban environments. The newly-built surface structures would result in an increase in the lining force. More importantly, during earthquake loading the surface buildings would alter the wave propagation, and complex dynamic interaction may occur between the metro tunnel and the adjacent structure. Hence, considering the coupling effects of surface building and earthquake loading is crucial in the seismic safety evaluation of underground structures.

To better understand the dynamic behavior of underground structures subjected to seismic loadings, some studies have been carried out to investigate their seismic responses affected by surface buildings. Wang et al. [18] examined the interaction effects between the embedded tunnel and surface structure through a series of shaking table tests and found that the seismic responses of the tunnel-soil-surface-building system were significantly influenced by the type of earthquake motions. Pitilakis and Tsinidis [19], [20] studied the impact of buildings on the seismic behavior of circular tunnels under severe earthquake excitation. A general increase of the tunnel responses was presented with the consideration of buildings. Abate and Massimino [21], [22] treated the tunnel, the surrounding soil and the surface structure as an integral system, providing insights into the effect of surface building on the dynamic performance of a horseshoe-section tunnel. Dashti et al. [23] investigated the influence of mid- and high-rise buildings on the seismic responses of nearby box tunnels through centrifuge tests. Additionally, seismic response of subway station influenced by the surface buildings was analyzed [24], [25]. In most previous studies, the locations of surface buildings were fixed directly above or at a specific distance to the axis of the underground structure, while the effects of the location on the seismic responses were not investigated. On a related subject, the effects of underground structures on the seismic response of nearby surface buildings were investigated by several studies [26], [27], [28].

Although the seismic responses of underground structures affected by the presence of above-ground structures have been investigated in some previous studies, the shield tunnel structure was usually treated as an equivalent integral ring, and the focus was mainly concentrated on the dynamic forces and overall deformation of the tunnel. However, a shield tunnel is an assembled structural system with a large number of segment joints, which are the vulnerable elements of the structure. Hence, the seismic behavior of a shield tunnel lining may be quite different from that of a continuous lining. During seismic ground shaking, the tunnel does not only exhibit overall deformation but also openings and dislocations in segment joints, and the contact stress between two adjacent segments and the bolt forces will increase sharply, which should be closely investigated for the seismic safety of the tunnel structure. In addition, the influences of the relative positions between the surface and underground structures on the seismic responses should also raise attention.

The present study aims to qualitatively evaluating the transverse dynamic responses of in-service shield tunnels, considering the coupling effects of later-built surface structures and earthquake loading. To this end, a plane-strain numerical study employing the OpenSees platform is conducted in which the segment joints as well as their characteristics are taken into consideration, and a frame building at the ground surface is modeled after the ground gravity loading has been applied on the tunnel structure. The coupling effects are estimated through the ovaling distortion, dynamic lining force, joint displacement and dynamic internal forces of joint bolts. Parametric studies are carried out to investigate some influencing factors. Additionally, numerical analyses in which the tunnel lining is simplified as an equivalent ring by applying transverse stiffness reduction are also conducted.